JP2009529187A - Assessment of road traffic conditions using data from mobile data sources - Google Patents

Abstract

Techniques are described for assessing road traffic conditions in various ways based on obtained traffic-related data, such as data samples from vehicles and other mobile data sources traveling on the roads, as well as in some situations data from one or more other sources (such as physical sensors near to or embedded in the roads). The assessment of road traffic conditions based on obtained data samples may include various filtering and/or conditioning of the data samples, and various inferences and probabilistic determinations of traffic-related characteristics of interest from the data samples. In some situations, the inferences include repeatedly determining traffic flow characteristics for road segments of interest during time periods of interest, such as to determine average traffic speed, traffic volume and/or occupancy, and include weighting various data samples in various ways (e.g., based on a latency of the data samples and/or a source of the data samples).

Description

  The following disclosures are generally obtained from various data sources, such as by inferring traffic-related information for those roads based on data samples that reflect actual travel on the roads of interest. The present invention relates to a technique for evaluating road traffic conditions based on the collected data.

  As road traffic has continued to increase at a faster rate than the increase in road capacity, the impact of increasing traffic congestion has had an increasing detrimental impact on company and government activities and individual comfort. Accordingly, efforts have been made to combat the increasing traffic congestion in various ways, such as by obtaining information about current traffic conditions and providing that information to individuals and organizations. Such current traffic status information can be provided in various ways to interested parties (eg, frequent radio broadcasts, color for current traffic jams on several major roads in a geographical area) Via Internet websites, cell phones and other portable consumer devices that display maps of geographic areas with coded information).

US patent application Ser. No. 11 / 367,463

  One source for obtaining information about current traffic conditions is from observations supplied by humans (eg, traffic helicopters that provide general information about traffic flows and accidents, drivers via cell phones) But other sources in some larger metropolitan areas can measure the traffic flow of various roads in that area (eg, via sensors embedded in the road pavement) It is a network of traffic sensors. Human-supplied observations provide value in a limited situation, but such information is usually limited to only a few areas at a certain time and usually lacks enough detail to be very useful Yes.

  The traffic sensor network can provide more detailed information about traffic conditions on some roads in some situations. However, various problems exist with respect to such information as well as information provided by other similar sources. For example, many roads do not have road sensors (eg, are not large enough to have road sensors as part of a geographic area and / or nearby network that does not have a network of road sensors) Highways), even roads with road sensors, often do not provide accurate data, which greatly reduces the value of the data provided by traffic sensors. One source of inaccurate and / or unreliable data is a traffic sensor that breaks and thus provides no data, or provides intermittent or inaccurate data reading. Another source of inaccurate and / or unreliable data includes problems with temporal transmission of data from one or more sensors that result in intermittent delivery, delayed delivery, or no delivery of data. Furthermore, many traffic sensors are not configured or designed to report information about their operational status (eg, whether they are functioning correctly), even when operational status information is reported. , To determine if the information is inaccurate (eg, reporting that it is functioning normally when it isn't actually) and therefore the data provided by the traffic sensor is accurate Make it difficult or impossible. Furthermore, some traffic related information may be available only in raw and / or unaggregated form, and thus may have limited utility.

  Thus, it would be beneficial to provide various additional related functions while providing improved techniques for obtaining and evaluating traffic related information.

  Traffic obtained, such as data samples from vehicles moving on the road and other mobile data sources and / or from road traffic sensors (eg, physical sensors embedded in the road or otherwise close to the road) Describe techniques for evaluating road traffic conditions in various ways based on relevant data. Further, in at least some embodiments, one or more other data samples, such as by obtaining data samples from mobile data sources from physical sensors located near or embedded in the road Can be supplemented with data from other sources. Evaluation of road traffic conditions based on obtained data samples (eg, data reading from road traffic sensors, individual or aggregated data points from mobile data sources, etc.) includes various filtering and data sample and reading Various inferences and probabilistic determinations of traffic-related characteristics of interest and / or conditioning may be included.

  As noted, in some embodiments, the obtained road traffic situation data includes multiple data samples supplied by a mobile data source (eg, a vehicle), road-based traffic sensors (eg, road pavement). Read data from loop sensors embedded within) and data from other data sources. This data can be analyzed in various ways to facilitate the determination of traffic condition characteristics of interest, such as the estimated average traffic speed and the estimated total vehicle volume for a particular portion of the road of interest, Such traffic situation determinations can be performed in real-time or near real-time form (eg, within 2-3 minutes of receipt of underlying data samples and / or readings). For example, the acquired data can be conditioned in various ways to detect and / or correct errors in the data. If the obtained road traffic situation information data is further filtered in different ways in different embodiments and the data is inaccurate or does not represent the actual traffic situation characteristics that are of interest Identifying data samples that are statistically outliers with respect to data samples that are not of interest and / or other data samples based at least in part on the road with which the data samples are associated In some embodiments, filtering may further include performing an association of data samples with a particular road. The data sample to be filtered further includes a data sample that otherwise reflects a vehicle position or vehicle activity that is not of interest (eg, a parked vehicle, a parking lot or a vehicle turning in a parking building), and Data samples may be included that do not otherwise represent actual vehicle movement on the road of interest. The evaluation of the obtained data is at least in some embodiments based on traffic conditions (e.g., traffic flows and various parts of the road network within a particular geographic area based at least in part on the obtained data samples. (Or average traffic speed) determination may be included. The evaluated data can then be utilized to perform other functions related to the analysis, prediction, forecasting, and / or provision of traffic related information. In at least some embodiments, the data sample manager system uses a plurality of traffic conditions at a plurality of future times, as described in more detail below, using at least some of the described techniques. Prepare data to be used by a traffic data client, such as a predictive traffic information provider system that generates predictions for

  In some embodiments, conditioning the obtained data samples detects and / or corrects errors present in the data in various ways (eg, for data readings received from road traffic sensors). It may include correcting the data sample in error, such as by chance. Specifically, to determine whether these data sources are working correctly and provide reliable data samples reliably, such as based on analysis of data samples provided by specific data sources A technique for evaluating the “health” of these data sources (eg, road-based traffic sensors) is described. For example, in some embodiments, current traffic data readings may be caused by incorrect operation of traffic sensors or other problems in the data and / or otherwise abnormal The current data reading supplied by a given traffic sensor is used to determine whether it is significantly different from normal past data readings, such as those that may reflect current traffic conditions. Can be compared with past data readings (eg, historical average data) supplied by. Such detection and analysis of possible errors for a particular data source and / or within the current traffic data reading uses neural networks, Bayesian classifiers, decision trees, etc., as described in more detail below. It can be implemented in various embodiments in a variety of ways, including based at least in part on a classification technique, such as by chance.

  After detecting unreliable data samples, such as those from broken data sources that are operating incorrectly, such unreliable data samples (as well as missing data samples) can be corrected in various ways or otherwise Can be modified in the form. For example, missing and unreliable data samples of one or more data sources (eg, traffic sensors) are, in some embodiments, nearby or otherwise related traffic sensors that are operating correctly. Via concurrent data samples from (e.g., by averaging data readings supplied by nearby traffic sensors), via prediction information associated with missing and unreliable data samples (e.g. One or more data sources (by using predicted traffic information and / or forecast traffic information for one or more data sources to determine the expected data read of these data sources) Via historical information (for example, historical One or more other relevant information, such as through adjustments to incorrect data samples using consistent bias or other types of error information that causes compensable errors (by utilizing average data reading) It can be corrected by using the source. Additional details regarding the correction of missing and unreliable data samples are included below.

  In addition, traffic status information can be used in a variety of other ways, such as when data currently available may not be able to reliably modify data samples for a particular data source (eg, a particular traffic sensor). The technique for estimation will be described. For example, the presence of multiple nearby unhealthy traffic sensors that are operating incorrectly can result in insufficient data to evaluate traffic flow with sufficient confidence for each one of these traffic sensors. May cause a condition. In that case, the traffic situation information can be estimated in a variety of other ways, including based on a group of related traffic sensors and / or other information related to the structure of the road network. For example, as described in more detail below, each road of interest can be modeled or represented by the use of multiple road segments, each of which includes a plurality of associated traffic sensors. And / or have available data from one or more other data sources (eg, mobile data sources). If so, traffic status information evaluated for adjacent road segments, predicted information for a particular road segment (eg, based at least in part on the current state and recent state at the time of prediction, 3 Information generated for a limited future time period, such as time), forecast information for a specific road segment (for example, some or all of the current state information and recent state information used for forecasting are not used) Information generated for a longer future time period, such as 2 weeks or more), using a historical average state of a specific road segment, etc. Road traffic situation information can be estimated for other groups of traffic sensors). By utilizing such techniques, traffic status information may be provided even when there is little or no current traffic status data with respect to one or more nearby traffic sensors or other data sources. it can. Additional details regarding such traffic situation information estimation are included below.

  As previously noted, information regarding road traffic conditions can be obtained from mobile data sources in various forms and in various forms. In at least some embodiments, the mobile data source includes vehicles on the road, each of which includes one or more computing systems that provide data regarding the movement of the vehicle. For example, each vehicle may have a GPS (“Global Positioning System”) that can determine the geographic location, speed, direction, and / or other data that characterizes the vehicle's movement or otherwise relates to the vehicle's movement. ) Devices and / or other geolocation devices, and one or more devices on the vehicle (either geolocation device (s) or separate communication devices) are sometimes Such data can be provided (eg, via a wireless link) to one or more systems (eg, a data sample manager system described in more detail below) that can use the data. Such vehicles include, for example, distributed networks of vehicles driven by individual unrelated users, fleets of vehicles (eg, transportation companies, taxis and bus companies, transportation companies, government bodies or governments) For vehicles belonging to commercial networks (eg, OnStar services) that supply relevant information, such as agencies, vehicle rental service vehicles, etc., to obtain such traffic status information (eg, by moving along a predefined route) Or a group of vehicles to be driven to move by moving on a dynamically directed road, such as to obtain information about a road of interest, (eg, GPS capabilities of the device) Based on geolocation and / or provided by cellular network Zui it) on-board cellular telephone device (e.g., can include such as a vehicle having has as a and / or vehicle occupant embedded devices).

  In at least some embodiments, mobile data sources include or include computing devices and other mobile devices of users moving on the road, such as users who are drivers and / or passengers of vehicles on the road. It can be based on. Such user devices may include devices with GPS capabilities (eg, cell phones and other handheld devices), or other location information and / or movement information instead in other embodiments. Can be made in shape. For example, an external system in which a device in a vehicle and / or a user device can detect and track information about the device (eg, for a device passing through each of a plurality of transmitters / receivers in a network operated by the system) Can thus determine the location information and / or movement information of the device in various ways at various levels of detail, or such an external system in other ways, In some cases, information about the vehicle and / or user can be detected and tracked without interacting with the device (eg, a camera system that can observe and identify the license plate and / or the user's face). Such external systems include, for example, cell phone towers and cell phone networks, other wireless networks (eg, Wi-Fi hotspot networks), vehicle transponder detectors (eg, RFID) that use various communication techniques. That is, “Radio Frequency Identification”), vehicle and / or other detectors of the user (eg, infrared distance measuring device, sonar distance measuring device, radar distance measuring device, or laser distance measuring device) Or determine the speed).

  Road traffic status information obtained from mobile data sources can vary widely, either alone or in combination with other road traffic status information from one or more other sources (eg, from road traffic sensors). Can be used in form. In some embodiments, such road traffic situation information obtained from a mobile data source is similar to information from road sensors, but information about roads that do not have a functioning road sensor (eg, a network of road sensors). No geographic areas and / or for roads that are not so large as to have road sensors, such as for roads lacking sensors, for broken road sensors, etc.), received from road sensors or other sources Used to verify duplicated information, identify road sensors that are providing inaccurate data (eg, due to time issues or ongoing issues), and the like. In addition, an absolute term (e.g. one to indicate the average speed, the amount of traffic for a given period of time, the average percentage of time that the vehicle is on the sensor or otherwise activated the sensor, etc. or One of a plurality of listed levels of road congestion, such as an average occupancy time of multiple traffic sensors or other locations on the road, one measured based on one or more other traffic condition measurements Etc.) and / or relative terms (e.g. to represent differences from normal or maximum values), etc., whether based on data samples from mobile data sources and / or from traffic sensor data reading Road traffic conditions can be measured and represented in one or more of various forms.

  In some embodiments, some road traffic situation information can take the form of data samples provided by various data sources, such as data sources associated with the vehicle, to report vehicle movement characteristics. . Individual data samples can contain varying amounts of information. For example, a data sample provided by a mobile data source can include one or more of a source identifier, a speed indication, a direction or direction indication, a location indication, a time stamp, and a status indicator. The source identifier may be a number or string that identifies the vehicle (or person or other device) that serves as the mobile data source. In some embodiments, the mobile data source identifier is permanently or temporarily (e.g., every time the vehicle or data source device is powered on for an hour during the lifetime of the mobile data source). Can be associated with a mobile data source (eg during the current session of use). In at least some embodiments, the source identifier is mobile data, such as by creating and / or manipulating the source identifier in a manner that prevents a mobile data source associated with the identifier from being identified based on the identifier. Associated with a mobile data source (whether associated permanently or temporarily) in a manner that minimizes privacy concerns associated with data from the source. The speed display can reflect the instantaneous or average speed of the mobile data source expressed in various ways (eg, miles per hour). The direction of travel can reflect the direction of travel and can be an angle expressed in degrees or other units of measurement (eg, compass-based direction of travel or radians). The display of the location can reflect the physical location expressed in various forms (eg, latitude / longitude pairs or universal transverse Mercator coordinates). The time stamp may represent the time at which a given data sample was recorded by the mobile data source, such as local time or UTC (“Universal Standard Time”) time. The status indicator may indicate the status of the mobile data source (eg, the vehicle is moving, stopped, stopped with the engine running, etc.) and / or sensing device, recording device, and / or transmitting device. At least some of the situations (eg, low battery, low signal strength, etc.) can be indicated.

  In some embodiments, a network of roads within a given geographic region can be modeled or represented by the use of multiple road segments. Each road segment is by dividing a given physical road into a plurality of road segments (eg, each road segment is a specific length, such as a road that is 1.6 km (1 mile) long, or a road segment Can be used to represent a part of a road (or multiple roads), such as with multiple selected roads that share similar traffic conditions characteristics) The road segments may be contiguous portions of the road, or alternatively, in some embodiments, may include overlapping or intervening road portions that are not part of any road segment. Can have. Further, a road segment can represent one or more lanes of travel on a given physical road. Thus, a particular multi-lane road having one or more lanes of movement in each of the two directions can be associated with at least two road segments, and at least one road segment is associated with movement in one direction And at least one other road segment is associated with movement in the other direction. Further, multiple lanes of a single road for movement in a single direction can be represented by multiple road segments in some situations, such as when the lanes have different movement condition characteristics. For example, a given freeway system is represented by a road segment that is distinct from a road segment that represents a normal (eg, non-HOV) lane that travels in the same direction as an express lane or carpool-only (“HOV”) lane. May have an express lane or HOV lane that may be beneficial. A road segment can be connected or otherwise associated with other nearby road segments, thereby forming a network of road segments.

  FIG. 1 is a block diagram illustrating the data flow between components of an embodiment of a data sample manager system. The illustrated data flow diagram is intended to reflect a logical representation of the data flow between the data source, the components of the data sample manager system embodiment, and the traffic data client. That is, the actual data flow can be a direct flow (eg, implemented by network communication such as parameter passing or messages) and / or an indirect flow through other storage mechanisms such as one or more database systems or file systems. It can occur through a variety of mechanisms. The illustrated data sample manager system 100 includes a data sample filterer component 104, a sensor data conditioner component 105, a data sample outlier eliminator component 106, a data sample rate assessor component 107, a data sample flow assessor component 108, and options. A sensor data aggregator component 110.

  In the illustrated embodiment, components 104-108 and 110 of data sample manager system 100 obtain data samples from a variety of data sources including vehicle-based data source 101, road traffic sensor 103, and other data sources 102. . The vehicle-based data source 101 can include a plurality of vehicles traveling on one or more roads, each of which supplies one or more data relating to the movement of the vehicle. Computing systems and / or other devices may be included. As described in greater detail elsewhere, each vehicle can include a GPS and / or other geolocation device that can determine position, speed, and / or other data associated with the movement of the vehicle. Such data may be transmitted to the vehicle by a wireless data link (eg, satellite uplink and / or cellular network) or otherwise (eg, when a fleet vehicle returns to the home base). (Via connections with physical wires / cables made when arriving at a location) can be obtained by the components of the data sample manager system described. The road traffic sensor 102 includes a loop sensor embedded in the pavement that can measure the number of vehicles passing over the sensor per unit time, vehicle speed, and / or other data related to traffic flow, etc. Multiple sensors installed in or near various streets, highways, or other roads can be included. Data can also be obtained from the road traffic sensor 102 via a wired or wireless based data link. Other data sources 103 include information regarding the road network, such as connections between roads, and traffic control information associated with such roads (eg, presence and / or location of traffic control signals and / or speed limit segments). Various other types of data sources may be included, including a map service to provide and / or a map database.

  The illustrated data sources 101-103 in this example provide data samples directly to the various components 104-108 and 110 of the data sample manager system 100, but instead, in other embodiments, It can be processed in various ways before supply to these components. Such processing includes organizing and / or aggregating data samples into logical collections based on time, location, geographic region, and / or identity of individual data sources (eg, vehicles, traffic sensors, etc.). Can do. Further, such processing may include merging data samples into higher order logical data samples or other values or otherwise combining them. For example, data samples obtained from multiple geographically co-located road traffic sensors can be merged into a single logical data sample by averaging or other aggregation. Further, such processing may include derivation of data samples or elements of data samples based on one or more obtained data samples or other forms of synthesis. For example, in some embodiments, at least some vehicle-based data sources can each provide a data sample that includes only a source identifier and a geographic location, and if so, a particular time interval Or a group of multiple distinct data samples that are supplied periodically over other time periods can thereby be associated with each other as supplied by a particular vehicle. Such groups of data samples are then headed for each data sample (eg, by calculating the angle between the position of one data sample and the position of the previous and / or subsequent data sample. ) And / or velocity for each data sample (eg, by calculating the distance between the position of one data sample and the position of the previous and / or subsequent data sample and dividing this distance by the corresponding time), etc. Further processing related information can be determined to determine.

  The data sample filter component 104 obtains data samples from the vehicle-based data source 101 and other data sources 102 in the illustrated embodiment, and then passes the obtained data samples to the data sample outlier eliminator component 106 and optionally These data samples are filtered before being supplied to the data sample flow assessor component 108. As described in more detail elsewhere, such filtering may involve associating data samples with road segments that correspond to roads in a geographic area and / or data samples that do not correspond to road segments of interest or Identifying data samples that otherwise reflect the location or activity of vehicles that are not of interest can be included. For associating data samples to road segments, use the reported position and / or direction of each data sample to determine whether the position and direction of travel correspond to a previously defined road segment Can be included. Identification of data samples that do not correspond to the road segment of interest can be accomplished by removing or otherwise identifying such data samples so that the data samples can be Such data samples that should not be modeled, considered, or otherwise processed by the component and that should be removed include certain functional road types that are not of interest ( functional road classes) (for example, residential streets), data samples for specific uninterested roads or road segments, uninterested road sections or sections ( For example, freeway lamps and collector / distributors (colle tor / distributor) such as data samples corresponding to the lane / road) can be included. To identify data samples that otherwise reflect the location or activity of vehicles that are not of interest, you can drive an idle vehicle (for example, parked with the engine running) in a parking building. Identifying data samples corresponding to a vehicle (eg, turning at a very low speed) or the like. Further, filtering may include identifying road segments that are (or are not) interested in presentation or further analysis in some embodiments. For example, such filtering may include some or all road segments that have low intra-period variability and / or low congestion (eg, sensor data reading is not available or functional road type is otherwise Certain time periods (for example, hours), such as excluding smaller roads or road segments that show less moving roads from further analysis as having less interest than other roads and road segments , Day, week) analysis of traffic flow and / or congestion level variability of various road segments.

  The sensor data conditioner component 105 assists in correcting erroneous data samples, such as by detecting and correcting reading errors obtained from the road traffic sensor 103. In some embodiments, data samples that are detected by the sensor data conditioner component as unreliable are not forwarded for use to other components, such as the data sample outlier eliminator component 106 (or other components In order to be able to process these data samples accordingly, an indication that a particular data sample is unreliable is provided). If so, the data sample outlier eliminator component can determine if a sufficiently reliable data sample is available, and otherwise can initiate corrective action. Alternatively, in some embodiments and situations, the sensor data conditioner component further performs at least some corrections to the data samples as described in more detail below, after which the corrected data is converted to sensor data. The aggregator component 110 (and optionally other components such as a data sample outlier eliminator component and / or a data sample flow assessor component) may be provided. The detection of an erroneous data sample is reported by the road traffic sensor during the corresponding time period (eg, the same day and time) the distribution of the current data sample reported by the given road traffic sensor Various techniques can be used, including statistical measurements that are compared to the historical distribution of data samples. The degree to which the actual and historical distributions differ is due to statistical measurements such as the Kullback-Leibler information amount that provides a convex measure of similarity between two probability distributions and / or statistical information entropy. Can be calculated by: In addition, some road sensors can report sensor health indications, and such indications can also be used to detect errors in the acquired data samples. If errors are detected in the acquired data samples, the average of adjacent (eg, adjacent) data samples from adjacent / adjacent road sensors where such data samples have not been determined to be in error Errored data samples can be corrected in a variety of ways, including by substituting values. In addition, erroneous data samples may instead be corrected by using previously and / or previously predicted and / or predicted values, such as those that can be provided by a predictive traffic information system. Additional details regarding the predictive traffic information system are provided elsewhere.

  The data sample outlier eliminator component 106 obtains filtered data samples from the data sample filter component 104 and / or conditioned or otherwise modified data samples from the sensor data conditioner component 105; Thereafter, data samples that do not represent actual vehicle movement on the roads and road segments of interest are identified and excluded from consideration. In the illustrated embodiment, for each road segment that is of interest, this component uses a specific time period to determine which ones should be excluded if there are ones that must be excluded. Analyze groups of data samples recorded in and associated with road segments (eg, by data sample filter component 104). Such a determination of data samples not represented can be performed in a variety of ways, including based on techniques that detect data samples that are statistical outliers with respect to other data samples in the group of data samples. Additional details regarding data sample outlier removal are provided elsewhere.

  The data sample speed assessor component 107 is a data sample from the data sample outlier eliminator component 106 such that the obtained data sample of the illustrated embodiment represents actual vehicle movement on the roads and road segments of interest. Get The data sample speed assessor component 107 then analyzes the acquired data samples and associated with the road segment (eg, by the data sample filter component 104 or by reading from a traffic sensor that is part of the road segment. ) Evaluating one or more velocities of the interested road segment for at least one time period of interest based on the group of data samples and the time period. In some embodiments, the rate (s) evaluated may include one or more attributes of the data sample (eg, age such as to give a newer data sample greater weight, and / or By the source or type of data sample, such as to change the weight of a data sample from a mobile data source or road sensor to give greater weight to a source with higher expected reliability or availability) or by other factors A weighted average of the velocities of multiple data samples in the group can be included. Further details regarding rate estimation from data samples are provided elsewhere.

  Data sample flow assessor component 108 evaluates traffic (expressed as the total number or average number of vehicles arriving at or traversing a road segment over a specific length of time, such as every minute or hour, for example). In order to evaluate traffic density (expressed as the average number or total number of vehicles per unit of distance, eg every mile or kilometer), traffic occupancy (eg every minute or every hour etc. Interested in at least one time period of interest (e.g., expressed as an average or total amount of time a vehicle occupies a particular point or region over a particular length of time) The traffic flow information of the road segmentThe evaluation of traffic flow information in the illustrated embodiment is based at least in part on the traffic speed related information provided by the data sample speed assessor component 107 and the data sample outlier eliminator component 106, and optionally the sensor data conditioner component 105. And based on traffic data sample information provided by the data sample filter component 104. Additional details regarding data sample flow evaluation are provided elsewhere.

  If present, the sensor data aggregator component 110 may remove all unreliable data samples and / or correct all missing and / or unreliable data samples, such as after the sensor data conditioner component It aggregates sensor-based traffic situation information supplied by the sensor data conditioner component 105. In the alternative, the sensor data aggregator component can instead perform all such removal and / or correction of missing and / or unreliable data samples. In some cases, the sensor data aggregator component 110 aggregates (e.g., averages) information provided by a plurality of individual traffic sensors associated with each of the various road segments. Each traffic flow information can be provided. Accordingly, when present, the sensor data aggregator component 110 provides information complementary to the estimated traffic situation information provided by components such as the data sample rate assessor component 107 and / or the data sample flow assessor component 108. Or alternatively, no data samples from the mobile data source are available, or other components such as the data sample velocity assessor component 107 and the data sample flow assessor component 108 are accurately evaluated for road traffic conditions. It can be used when a sufficient amount of reliable data samples are not available to make it possible to provide information.

  The one or more traffic data clients 109 of the illustrated embodiment may include estimated road traffic situation information (eg, speed and / or flow) provided by the data sample speed assessor component 107 and / or the data sample flow assessor component 108. Data), and such data can be used in various ways. For example, the traffic data client 109 may be real-time to a predictive traffic information provider system and / or end users and / or third party clients that utilize traffic status information to generate future traffic status predictions at multiple future times. Other components operated by an operator of the data sample manager system 100, such as a real-time (or near real-time) traffic condition information presentation system or a real-time (or near real-time) traffic condition information provider system that provides (or near real-time) traffic condition information And / or a traffic information system. In addition, the traffic data client 109 can include a computing system operated by a third party to provide traffic information services to customers. In addition, the one or more traffic data clients 109 may optionally have sufficient data for some situations (eg, the data sample rate assessor component and / or the data sample flow assessor component to perform an accurate assessment). Instead of or in addition to data from the Data Sample Rate Assessor component and / or Data Sample Flow Assessor component (when not available and / or when data from a vehicle-based data source or other source is not available) In any case, road traffic situation information supplied by the sensor data aggregator component 110 can be obtained.

  For purposes of illustration, some embodiments are described below in which specific types of road traffic situations are evaluated in a particular way, and such evaluated traffic information is used in a variety of particular ways. However, it will be appreciated that such road traffic situation assessments can be generated in other forms, using other types of input data in other embodiments, and the techniques described can be used in a variety of other situations. Thus, the invention is not limited to the exemplary details provided.

  2A-2E illustrate examples of road traffic situation assessments based on data obtained from vehicles and other mobile data sources, such as those that can be performed by the described data sample manager system embodiments. Specifically, FIG. 2A shows an example of data sample filtering of an example area 200 having a plurality of roads 201, 202, 203, and 204 and having a legend display 209 that indicates the north direction. In this example, road 202 is a road with restricted entrance and exit, such as a freeway or toll road, and has two separate groups 202a and 202b of lanes for moving vehicles in the west and east directions, respectively. The lane group 202a includes an HOV lane 202a2 and a plurality of other normal lanes 202a1, and the lane group 202b similarly includes an HOV lane 202b2 and a plurality of other normal lanes 202b1. The road 201 is a main road having two lanes 201a and 201b for moving vehicles in the south direction and the north direction, respectively. Road 201 passes over road 202 (e.g., via an elevated crossover or overpass), and road 204 connects an on-ramp (on -ramp). The road 203 is a local side road adjacent to the road 202.

  The road shown in FIG. 2A can be represented in various ways for use by the described data sample manager system. One or more road segments can be associated with each physical road, such as, for example, to associate northbound and southbound road segments with northbound lane 201b and southbound lane 201b, respectively. Similarly, at least one westbound road segment and at least one eastbound road segment may be associated with a westbound lane group 202a and an eastbound lane group 202b of road 202, respectively. For example, the portion of the east lane group 202b east of the road 201 may cause road traffic conditions that normally or often change between the road portions (eg, more heavily congested in the lane group 202b east of the road 201). Separate roads from the east-bound lane group 202b west of the road 201, such as based on the normal ramp-up of vehicles from the on-ramp 204 to the east lane group 202b of the road 201 It can be a segment. Further, if different lanes usually or often have different road traffic characteristics (eg due to the first road segment corresponding to lane 202b1 and different traffic conditions based on lanes sharing similar traffic characteristics) One or more lane groups can be broken down into multiple road segments, such as to represent any given portion of lane group 202b as a second road segment corresponding to HOV lane 202b2. In such a situation, only a single road segment can be used for such a lane group, but some data samples (eg, data samples corresponding to HOV lane 202b2) are used for the road traffic of that lane group. Used when evaluating the situation (data sample filter It can be excluded from the like) due to components and / or data sample outlier eliminator component. In the alternative, some embodiments allow multiple lanes on a given road even when the lane is used for movement in the opposite direction, such as when road traffic conditions are usually similar in both directions. For example, the side road 205a can have two opposing moving lanes, but can be represented by a single road segment. In at least some embodiments, a road segment is associated with geographic information (eg, physical dimensions and / or direction of travel (s)) and / or traffic related information (eg, speed limits), etc. It can be determined at least partially in various other ways.

  FIG. 2A further illustrates multiple mobile data sources (eg, illustrated) that move within area 200 during a particular time interval or other time period (eg, 1 minute, 5 minutes, 10 minutes, 15 minutes, etc.). A plurality of data samples 205a-k reported by a non-vehicle). Each of the data samples 205a-k is shown as an arrow indicating the direction of travel of the data sample as reported by one of a plurality of mobile data sources. The data samples 205a-k may differ from the actual position of the vehicle when the data samples were recorded (eg, position sensing due to incorrect reading or erroneous reading or used) Area 200 in a manner that reflects the reported location (eg, based on GPS readings, expressed in units of latitude and longitude), etc. (due to the degree of variability inherent in the mechanism) Is superimposed. For example, data sample 205g shows a position slightly north of road 202b, which may reflect a vehicle that has been stopped off the north side of lane 202b2 (eg, due to a mechanical malfunction) Alternatively, it may reflect the incorrect position of the vehicle that was actually moving eastward in lane 202b2 or another lane. In addition, a single mobile data source was reported by a single vehicle, both sample 205i and sample 205h, based on eastbound travel along road 202 during the time period (eg, every 5 minutes or It can be a source of multiple of the illustrated data samples, such as when reporting data samples every 15 minutes (via a single transmission containing multiple data samples of multiple previous time points). Further details regarding the storage and provision of multiple acquired data samples are included below.

  The described data sample manager system, in some embodiments, to map data samples to predefined road segments and / or to identify data samples that do not correspond to a road segment of interest, etc. Obtained data samples can be filtered. In some embodiments, the reported location of the data sample is within a predetermined distance (eg, 5 meters) from the location of the road and / or lane (s) corresponding to the road segment and its progression Associate a data sample with a road segment if the direction is within a predetermined angle (eg, plus 15 degrees or minus 15 degrees) from the direction of travel of the road and / or lane (s) corresponding to the road segment Can do. The road segments in the illustrated embodiment are associated with location-based information sufficient to make such a determination (e.g., road segment travel direction, road segment physical boundaries, etc.), but in other embodiments, The association of data samples to road segments can be performed before the data samples are made available from the data sample manager system.

  As an illustrative example, data sample 205a can be associated with a road segment corresponding to road 203 because its reported location is within bounds on road 203 and its direction of travel is associated with road 203. This is because it is the same (or almost the same) as at least one of the traveling directions. In some embodiments, determining whether a data sample can be associated with a road segment when a single road segment is utilized to represent multiple lanes, some of which move in opposite directions To do so, the direction of travel of the data sample can be compared with the direction of travel of both road segments. For example, data sample 205k has a direction of travel that is substantially opposite to data sample 205a, but if a road segment corresponding to road 203 is used to represent two opposing lanes of road 203, data sample 205k. Can also be associated with the road segment.

  However, due to the proximity of the road 203 and the lane group 202a, the travel direction of the data sample 205k is the same (and almost the same) as the travel direction of the lane group 202a, so the reported position of the data sample 205k is The data sample 205k may also be able to reflect a moving vehicle within the lane group 202a, such as when within a margin of error regarding the position of a moving vehicle in one or more of the lanes 202a. is there. In some embodiments, such cases of multiple possible road segments of a data sample can be disambiguated based on other information related to the data sample, eg, in this case, Lane group 202a corresponds to a freeway with a 104 km / h (65 mph) speed limit, road 203 is a local side road with a 48 km / h (30 mph) speed limit, and the reported speed of the data sample is 120 km / h ( 75 mph) (e.g. if the association with the freeway lane (s) is much more likely than the association with the local side road) Can help in disambiguation. More generally, if the reported speed of data sample 205k is more similar to the observed or posted speed of road 203 than the observed or posted speed of lane group 202a. Such information can be used as part of the decision to associate a data sample with the road 203 rather than the lane group 202a. Instead, if the reported speed of data sample 205k is more similar to the observed speed or posted speed of lane group 202a than the observed speed or posted speed of road 203, It can be associated with the lane group 202a instead of 203. Use other types of information as part of such disambiguation, such as as part of a weighted analysis to reflect the degree of match with candidate road segments for each type of information in the data sample (E.g., location, direction of travel, status, information about other related data samples such as other recent data samples from the same mobile data source, etc.).

  For example, with respect to associating data sample 205b to the appropriate road segment, the reported location appears in the overlap between lane 201b and lane group 202a and is near lane 201a as well as other roads. However, the reported direction of travel of this data sample (approximately northbound) matches the direction of travel of lane 201b (northbound) much better than the direction of travel of the other candidate lanes / roads, so this data sample is In this example, there is a high possibility that the road segment corresponds to the lane 201b. Similarly, data sample 205c includes a reported location that may match multiple roads / lanes (eg, lane 201a, lane 201b, and lane group 202a), but in its direction of travel (almost westbound). Can be used to select the road segment of lane group 202a as the most appropriate road segment of the data sample.

  Continuing with this example, data sample 205d has its direction of travel (almost eastward) because its position is opposite to the direction of travel of lane group 202a (westbound) corresponding to the reported position of this data sample, Cannot be associated with any road segment. If there are no other suitable candidate road segments close enough (eg, within a predetermined distance) to the reported location of the data sample 205d, such as when a lane group 202b with a similar direction of travel is too far, this data The sample can be excluded from subsequent use in the analysis of the data sample during filtering.

  Where the position-based technique used for the location of the data sample has sufficient resolution to distinguish between lanes (eg, differential GPS, infrared distance measuring device, sonar distance measuring device, or radar distance measuring device) Since the reported position and travel direction of the data sample 205e corresponds to the position and travel direction of the lane group 202a, the data segment 205e is a road segment corresponding to the lane group 202a, such as a road segment corresponding to the HOV lane 202a2. Can be associated with Data samples can also be associated with a particular lane of a multi-lane road based on factors other than location-based information, such as when the lanes have different traffic situation characteristics. For example, in some embodiments, using the reported rate of data samples, the expected distribution of the observed rate of data samples per candidate lane (or other measure of traffic flow) (eg, By modeling the normal distribution or Gaussian distribution and determining the best fit of the data sample to the expected distribution, the data sample can be fitted or matched to a particular lane. For example, based on analysis of other data samples (eg, using data readings supplied by one or more road traffic sensors) and / or other relevant current data, The reported speed of the data sample, such as by determining the inferred average speed, is greater than the observed average speed, inferred average speed, or historical average speed of a vehicle traveling in the lane 202a1 in the HOV lane 202a2. Since it is close to the observed average speed, inferred average speed, or historical average speed of a vehicle traveling within, data sample 205e can be associated with the road segment corresponding to HOV lane 202a2.

  In a similar manner, data samples 205f, 205h, 205i, and 205j can be associated with road segments corresponding to lane 201a, lane 202b1, lane 202b1, and ramp 204, respectively, because these data samples are reported. This is because the position and traveling direction correspond to the position and traveling direction of these roads or lanes.

  Data sample 205g can be associated with a road segment corresponding to lane group 202b (eg, a road segment in HOV lane 202b2) even if its reported location is outside the illustrated road boundary. This is because the reported location may be within a predetermined distance (eg, 5 meters) from the road. Instead, data sample 205g may not be associated with any road segment if its reported location is sufficiently far from the road. In some embodiments, different predetermined distances can be used for data samples provided by different data sources, such as to reflect the known or expected level of accuracy of the data source. For example, data samples provided by a mobile data source that utilizes uncorrected GPS signals can use a predetermined distance that is relatively long (eg, 30 meters), but mobile that utilizes a GPS device that is differentially corrected. Data samples provided by the data source can be compared using a relatively short (eg, 1 meter) predetermined distance.

  In addition, data sample filtering can include identifying data samples that do not correspond to a road segment of interest and / or that do not represent actual vehicle movement on the road. For example, some data samples can be removed from consideration because they are associated with roads that are not being considered by the data sample manager system. For example, in some embodiments, data samples associated with roads of lower functional road classes (eg, residential streets and / or trunk lines) can be filtered. Referring back to FIG. 2A, data sample 205a and / or 205k can be filtered because road 203 is a local road with a sufficiently low functional classification that is not considered by the data sample manager system, or data sample 205j is This on-ramp is separated from the freeway and is too short to be of interest so it can be filtered. Filtering may also include other mobile data source inferred activities or reported activities relative to other mobile data source inferred activities or reported activities on one or more road segments, etc. It can be based on factors. For example, a series of data samples associated with a road segment and supplied by a single mobile data source, all showing the same location, is likely to indicate that the mobile data source is down. If all other data samples related to the same road segment indicate a moving mobile data source, the data sample corresponding to the stopped mobile data source is due to the mobile data source being a parked vehicle, etc. It can be filtered out as not representing actual vehicle movement on the road segment. Further, in some embodiments, the data sample is a reported indication of the driving status of the vehicle (e.g., the vehicle is stopped to perform delivery, such as a vehicle transmission is in a “parked” state with the engine running). Such a display can be used as well to filter such data samples as not representing an actual moving vehicle. .

  FIG. 2B illustrates a plurality of data sources during a particular time interval or other time period in which data samples are plotted on a graph 210 where time is measured on the x-axis 210b and velocity is measured on the y-axis 210a. 2 shows a graphical view of a plurality of data samples associated with a single road segment obtained from. In this example, the data samples shown are obtained from multiple mobile data sources associated with the road segment as well as one or more road traffic sensors and differ as shown in the displayed legend. The shape is shown (ie, the filled diamond (“◆”) of the data sample obtained from the road traffic sensor) and the open square (“□”) of the data sample obtained from the mobile data source. make use of). Data samples from the illustrated mobile data source may be associated with a road segment as described with reference to FIG. 2A.

  Exemplary data samples include road traffic sensor data samples 211a-c and mobile data source data samples 212a-d. The reported speed and recording time of a given data sample can be determined by its position on the graph. For example, the mobile data source data sample 212d has a reported speed of 24 km per hour (15 miles per hour) (or other speed units) and is approximately 37 minutes relative to a starting point (or other time). Unit) time. As described in more detail below, some embodiments analyze or otherwise process acquired data samples within a particular time window during the represented time period, such as time window 213. can do. In this example, the time window 213 includes data samples recorded during a 10 minute interval from time 30 minutes to time 40 minutes. Further, some embodiments may further partition a group of data samples that appear within a particular time window into a plurality of groups, such as group 214a and group 214b. For example, the data sample shown appears to reflect a reported bimodal distribution of velocities, with a population of data samples in the range of 40-48 km (25-30 miles) per hour and 0-12.8 km per hour ( Note that speeds in the range of 0-8 miles are reported. Such bimodal distributions and other multimodal distributions of speed are, for example, due to traffic control that causes the traffic to flow in a lazy driving pattern, or multiple lanes of traffic traveling at different speeds (eg, other This can occur because the underlying traffic flow pattern is non-uniform, such as due to road segments that include HOV lanes or express lanes that have a relatively higher speed than non-HOV lanes. When such a multimodal distribution of speed data exists, some embodiments may provide improved accuracy or resolution of processing (eg, discrete average speeds that more accurately reflect the speed of various traffic flows). As well as to make additional information of interest (eg, speed differences between HOV and non-HOV traffic) or to identify groups of data samples to be excluded (eg, Data samples can be divided into groups for further processing, such as to avoid including HOV traffic as part of subsequent analysis. Although not shown herein, such separate groups of data samples may take various forms, including by modeling a separate distribution (eg, normal distribution or Gaussian distribution) of the observed velocity of each group. Can be identified.

  FIG. 2C is based on the reported rate of the data sample in this example (in other embodiments, either instead of or in addition to the reported rate, one of the data samples or Data sample outliers for filtering or otherwise excluding data samples that do not represent vehicles traveling on a particular road segment (although several other attributes can be used instead) An example of performing removal is shown. Specifically, FIG. 2C shows data performed for an example group of 10 data samples (in actual use, the number of data samples analyzed may be much larger). A table 220 showing sample outlier removal is shown. The illustrated data samples can be, for example, all of the data samples that occur within a particular time window (such as time window 213 in FIG. 2B), or alternatively, only a subset of the data samples in a particular time window. (Such as those included in group 214a or 214b in FIG. 2B) or all data samples available for a longer time period.

  In the current example, the data samples that are not represented were determined for the data samples by determining the deviation of the speed of each data sample in a group of data samples from the average speed of other data samples in the group. Identified as being statistical outliers for other data samples in the group. The deviation of each data sample can be measured, for example, with respect to the number of standard deviation differences from the average rate of the other data samples in the group, and the data sample whose deviation exceeds a predetermined threshold (eg, 2 standard deviations). Are identified as outliers and excluded from further processing (eg, by being discarded).

  The table 220 includes a heading row 222 that describes the contents of the plurality of columns 221a-f. Each row 223a-j of the table 220 shows a data sample outlier removal analysis of one of the 10 data samples, and column 221a shows the data sample analyzed for each row, where each data sample is analyzed. When done, that data sample is excluded from the other samples in the group to determine the resulting difference. The data sample in row 223a may be referred to as a first data sample, the data sample in row 223b may be referred to as a second data sample, and so on. Column 221b contains the reported speed of each of the data samples, measured in miles per hour. Column 221c lists the other data samples in the group to which the data samples in a given row are compared, and column 221d lists the approximate average rate of the group of data samples indicated by column 221c. Column 221e includes an approximate deviation between the rate of the excluded data sample from column 221b and the average rate listed in column 221d of other data samples, measured in units of standard deviation. Column 221f indicates whether a given data sample is removed based on whether the deviation listed in column 221e in this example is greater than 1.5 standard deviations. Further, the average speed 224 for all 10 data samples is shown to be about 41.12 km (25.7 miles) per hour, and the standard deviation 225 for all 10 data samples is shown to be about 14.2. ing.

  Thus, for example, row 223a indicates that the speed of data sample 1 is 41.6 km (26 miles) per hour. The average speed of the other data samples 2-10 is then calculated as approximately 41.12 km (25.7 miles) per hour. Next, the deviation of the speed of data sample 1 from the average speed of the other data samples 2-10 is approximately. Calculated as being 02 standard deviation. Finally, data sample 1 is determined not to be an abnormal value because its deviation is less than the threshold of 1.5 standard deviations. Furthermore, row 223c shows that the speed of data sample 3 is 0 km (0 mph) and the average speed of the other data samples 1-2 and 4-10 is approximately 45.76 km (28.6 mph). It is calculated as Next, the deviation of the speed of data sample 3 from the average speed of the other data samples 1-2 and 4-10 is calculated as approximately 2.44 standard deviations. Finally, since the deviation of the data sample 3 exceeds the threshold value of 1.5 standard deviation, it is determined that the data sample 3 is removed as an abnormal value.

More formally, given the N data samples ν ₀ , ν ₁ , ν ₂ ,..., Ν _n recorded during a given time period and associated with a given road segment, the current data sample ν _i is

Where ν _i is the speed of the current data sample being analyzed,

Is the average value of the velocities of the other data samples (ν ₀ ,..., Ν _i−1 , ν _{i + 1} ,..., Ν _n ), σ _i is the standard deviation of the other data samples, and c Is a constant threshold (eg, 1.5). In addition, as a special case for handling potential division by zero, the current sample ν _i has a standard deviation σ _i of other data samples of 0 and the current data sample speed is Average speed

If it is not equal to, it is removed.

Average value for each ν _i

Note that it is not necessary to iterate over all of the other data samples (ν ₀ ,..., Ν _i−1 , ν _{i + 1} ,..., Ν _n ) to calculate the standard deviation σ _i . Average value of other data samples ν ₀ , ..., ν _i-1 , ν _{i + 1} , ..., ν _n

Is

It can be expressed as, other data samples _{ν 0, ..., ν i-} 1, ν i + 1, ..., a standard deviation sigma _i of [nu _n is

Where N is the total number of data samples (including the current data sample),

Is the average value of all the data samples ν ₀ , ν ₁ , ν ₂ ,..., Ν _n , ν _i is the current data sample, and σ is the data samples ν ₀ , ν ₁ , ν ₂ , ..., all the standard deviations of ν _n . By using the above formula, the average value and the standard deviation can be calculated efficiently. Specifically, it can be calculated in a certain time. Since the above algorithm calculates the mean and standard deviation for each data sample in each road segment, the algorithm operates in O (MN) time, where M is the number of road segments, N is the number of data samples per road segment.

  In other embodiments, a technique based on neural network classifiers, naive Bayesian classifiers, and / or regression modeling, and groups of multiple data samples, either instead of or in addition to the outlier detection described, are combined. Other outlier detection algorithms and / or data removal algorithms can be used, such as techniques considered in (e.g., where at least some data samples are not independent of other data samples).

  FIG. 2D shows an example of performing an average speed assessment using data samples, showing an example data sample similar to that shown in FIG. 2B for a particular road segment and period of time. Data samples are plotted on graph 230, time is measured on the x-axis 230b and velocity is measured on the y-axis 230a. In some embodiments, the average speed of a given road segment can be calculated periodically (eg, every 5 minutes). Each calculation can take into account multiple data samples within a predetermined time window (or interval), such as 10 minutes or 15 minutes. To take into account the age of data samples when the average speed is calculated across such time windows, such as at or near the end of the time window (eg, for changing traffic conditions) Due to the intuition and expectation that it will not provide as accurate information as the younger data samples recorded relatively close to the current time regarding the actual traffic situation at the end of the time window or other current time, When aggregating the speed of data samples (to discount older data samples), the data samples within the time window can be weighted in various ways. Similarly, in some embodiments, when weighting data samples, the type of data source or a particular data source of the data sample (eg, the data sample is more accurate or otherwise different than the other source). Other data such as one or more other types of weighting factors), as well as one or more other types of weighting factors) Sample attributes can be considered.

  In the example shown, the average speed of the example road segment is calculated every 5 minutes over a 15 minute time window. This example shows the relative weights of the two exemplary data samples 231a and 231b because they contribute to the calculated average speed of each of the two time windows 235a and 235b. Time window 235a includes data samples recorded between times 30 and 45, and time window 235b includes data samples recorded between times 35 and 50. Both data samples 231a and 231b are included in both time windows 235a and 235b.

  In the illustrated example, each data sample within a given time window is weighted in proportion to its age. That is, older data samples are weighted less than younger data samples (thus contributing less to the average rate). Specifically, the weight of a given data sample decreases exponentially with age in this example. This decaying weighting function is illustrated by two weighting graphs 232a and 232b corresponding to time windows 235a and 235b, respectively. Each of the weight graphs 232a and 232b is a plot of the data sample recording time on the x-axis (horizontal) against the weight on the y-axis (vertical). Samples recorded later in time (ie, closer to the end of the time window) are weighted more than samples recorded earlier in time (ie, closer to the beginning of the time window). . The weight of a given data sample can be visualized by dropping a vertical line from the data sample in graph 230 down to the intersection of the weight graph curve corresponding to the time window of interest. it can. For example, the weight graph 232a corresponds to the time window 235a, and according to the relative ages of the data samples 231a (older) and 231b (younger), the weight 233a of the data sample 231a is less than the weight 233b of the data sample 231b. Furthermore, it can be seen that the weight graph 232b corresponds to the time interval 235b, and the weight 234a of the data sample 231a is smaller than the weight 234b of the data sample 231b. Furthermore, it is clear that the weight of a given data sample decays over time with respect to the subsequent time window. For example, the weight 233b of the data sample 231b in the time window 235a is greater than the weight 234b of the same data sample 231b in the later time window 235b. This is because data sample 231b is relatively younger during time window 235a compared to time window 235b.

More formally, in one embodiment, the weight of the data sample recorded at time t with respect to the time ending at time T is
w (t) = e ^{−α (Tt)}
Where e is a well-known mathematical constant and α is a variable parameter (eg, 0.2). Given the above equation, N pieces of data samples [nu ₀ in time interval ending at time _{T, ν 1, ν 2,} ..., a weighted average rate of [nu _n can be expressed as the following equation, t _i is , The time represented by the data sample ν _i (eg, the time when it was recorded).

  Further, the calculated average speed error estimate can be calculated as follows.

Here, N is the number of data samples, sigma is the sample [nu ₀ from the average _{velocity, ν 1, ν 2, ...} , it is the standard deviation of the [nu _n. In other embodiments, other forms of confidence values can be similarly determined for the calculated average speed or the generated average speed.

  As noted, data samples can be weighted based on other factors, either instead of or in addition to the freshness of the data samples. For example, for data samples, as described above, but by utilizing a different weight function (eg, to reduce the weight of data samples linearly rather than exponentially with age) Can be weighted. Further, the data sample weighting can be further based on the total number of data samples within the time interval of interest. For example, to reflect the increased likelihood of having a shorter latency (eg, younger) data sample that can be used to calculate the average rate, a larger number of data samples than for older data samples The variable parameter α described above may depend on or otherwise vary based on the total number of data samples so as to provide a large penalty (eg, less weight). In addition, the data samples can be weighted based on other factors including the type of data source. For example, a particular data source (eg, a particular road traffic sensor or all traffic sensors of a particular network) is known to be unreliable or otherwise inaccurate (eg, a reported situation (Based on information) or expected (eg, based on historical observation). In such cases, data samples obtained from such road traffic sensors (eg, data sample 211a in FIG. 2B, etc.) can be replaced with data samples obtained from mobile data sources (eg, data sample 212a in FIG. ) Can be less weighted.

  FIG. 2E facilitates an example of performing traffic flow assessment for road segments based on data samples, such as those that can include inferring traffic volume, traffic density, and / or traffic occupancy. In this example, traffic for a given road segment is expressed as the total number of vehicles flowing through the road segment within a given time window or the total number of vehicles arriving at the road segment during that time window. The traffic density for a road segment is expressed as the total number of vehicles per unit distance (for example, miles or kilometers), and traffic occupancy is the average time a vehicle is occupied by a particular road segment or a point on that road segment Expressed as a length.

  Given the number of distinct mobile data sources observed to be traveling on a given road segment during a given time window and the known or expected percentage of the entire mobile data source vehicle For example, it is possible to deduce the total traffic, ie the total number of vehicles (including vehicles that are not mobile data sources) that move on the road segment during the time window. From the inferred total traffic volume and the estimated average speed of vehicles on the road segment, it is possible to further calculate traffic density and road occupancy.

  An unsophisticated approach to estimating the total traffic for a particular road segment during a particular time window simply simplifies the number of mobile data sample sources in that time window by the percentage of actual vehicles expected to be mobile data sample sources. Thus, for example, mobile data samples are received from 25 mobile data sources during the time window and 10% of the total vehicles on the road segment are expected to be mobile data sample sources If so, the estimated total amount will be 250 actual vehicles for the length of time in the time window. However, this approach can lead to large fluctuations in the amount estimates of adjacent time windows due to the inherent variability in the arrival rate of vehicles, especially when the expected percentage of mobile data sample sources is small. is there. As an alternative that provides a more sophisticated analysis, the total traffic for a given road segment can be inferred as follows. Given a certain number n of observations of separate mobile data sources (eg, individual vehicles) on a road segment of length l during a given period τ, Bayesian statistics are used to The average rate λ on which the arrival is based can be inferred. The arrival of a mobile data source in a stretch of road corresponding to a road segment can be modeled as a random discrete process over time and can therefore be described by Poisson statistics as follows.

  From the above equation, the likelihood that n mobile data sources will be observed can be calculated given the average arrival rate λ and the observed number n of vehicles. For example, assume an average arrival rate of λ = 10 (vehicle / unit time) and observation of n = 5 vehicles. By assignment

Which shows 3.8% likelihood of actually observing n = 5 vehicles. Similarly, when the average arrival rate is λ = 10 (vehicle / unit time), the likelihood of actually observing 10 vehicles (ie, n = 10) is about 12.5%. It is.

  The above equation can be used with Bayesian theory to determine the likelihood of a particular arrival rate λ given n observations. As we know, Bayesian theory

It is.

  By substitution and constant removal, the following equation can be obtained.

  From above, given the observation of n mobile data sources, the proportional or relative likelihood of arrival rate λ can be calculated, spanning the possible values of λ for the various observed values of n A probability distribution is provided. For a particular value of n, the likelihood distribution across various arrival rate values selects a single representative arrival rate value (eg, a mean or median) and evaluates the confidence in that value. Make it possible.

  Furthermore, given the known percentage q of total vehicles on the road that is the mobile data source, also called “dominance rate”, the total traffic arrival rate amount is

Can be calculated as

  The total traffic volume of a road segment during a period of time can in some embodiments be instead expressed as the total number k of vehicles flowing in time τ across the length l of the road segment.

  FIG. 2E shows the probability distribution of various total traffic versus observed sample size given an example mobile data source dominance rate of q = 0.014 (1.4%). Specifically, FIG. 2E shows the mobile data source on the y-axis 241 against the inferred traffic arrival rate quantity on the x-axis 242 and the likelihood of each inferred traffic value on the z-axis 243. A three-dimensional graph 240 in which the number of observed units (n) is plotted is shown. For example, the graph is given the observed number of mobile data sources with n = 0, and the likelihood that the actual traffic is close to 0, as shown by bar 244a, is about 0.6 (ie 60% ) And the likelihood that the actual traffic volume is close to 143 vehicles per unit time is about 0.1, as indicated by the bar 244b. Furthermore, given the observed number of mobile data sources of n = 28, the actual total traffic is close to 2143 vehicles per unit time (if given the example control rate, about 30 units per unit time) The likelihood (corresponding to the mobile data sample source) is about 0.1, as shown by bar 244c, which is likely to be close to the median actual total traffic.

  In addition, the average occupancy and density are calculated as the inferred total traffic arrival rate amount for a given road segment (representing the number k of vehicles arriving on that road segment during time τ), the estimated average speed ν, and the average Using the vehicle length d, it can be calculated as follows.

Occupation = md
As previously described, the average vehicle speed ν on the road segment can be obtained by utilizing a speed estimation technique such as that described with reference to FIG. 2D.

  FIGS. 10A-10B show examples of conditioning and other forms of correction of erroneous data samples from road traffic sensors, such as unreliable and missing data samples. Specifically, FIG. 10A shows multiple example data reads obtained from multiple traffic sensors at various times organized in table 1000. Table 1000 shows a plurality of data read rows 1004a-104y, each of which is reported by a traffic sensor ID ("identifier") 1002a, which uniquely identifies the traffic sensor supplying the read. Traffic sensor data read value 1002b including the traffic flow information, traffic sensor data read time 1002c reflecting the time when the data read was taken by the traffic sensor, and traffic sensor state 1002d including an indication of the operating state of the traffic sensor. . In this example, only speed information is shown, but in other embodiments, additional types of traffic flow information can be reported by traffic sensors (eg, traffic and occupancy) and values can be in other formats. Can be reported.

  In the illustrated example, the data readings 1004a to 1004y are taken and recorded by a plurality of traffic sensors at various times as shown in the table 1000. In some cases, data readings may be taken periodically by traffic sensors (eg, every 5 minutes, every minute, etc.) and / or reported periodically by such traffic sensors. it can. For example, the traffic sensor 123 may include a plurality of data taken by the traffic sensor 123 between 10:25 AM and 10:40 AM on two separate days (8/13/06 and 8/14/06 in this example). Data readings are taken every 5 minutes, as indicated by data readings 1004a to 1004d and 1004f to 1004i indicating reading.

  Each illustrated data read 1004a-1004y includes a data read value 1002b that includes traffic flow information observed by the data sensor or otherwise obtained. Such traffic flow information may include the speed of one or more vehicles traveling at or near the traffic sensor at the traffic sensor. For example, data readings 1004a to 1004d indicate that the traffic sensor 123 has 54.4 km (34 miles) per hour, 57.6 km (36 miles) per hour, 67.2 km (42 miles per hour), and hourly at four different times, respectively. Indicates that the vehicle speed of 60.8 km (38 miles) was observed. Further, the traffic flow information includes the total number of vehicles moving at or near the traffic sensor, either in place of or in addition to speed information and / or other information. A count or incremental count can be included. The total count may be the cumulative count of vehicles observed by that traffic sensor since the sensor was installed or otherwise activated. The incremental count can be a cumulative count of vehicles observed by the traffic sensor since the traffic sensor took the previous data reading. Data reads 1004w-10004x indicate that the traffic sensor 166 has counted 316 cars and 389 cars, respectively, at two different times. In some cases, such as when a given traffic sensor experiences sensor malfunctions and cannot be reported or recorded (eg due to a network failure), the recorded data read is May not contain readings. For example, the data read 1004k could not provide the data read value to the 10:25 AM on the 8/13/06 day as indicated by “-” in the data read value column 1002b. Indicates.

  Further, the traffic sensor state 1002d can be associated with at least some data readings, such as when the traffic sensor and / or the corresponding communication network provides an indication of traffic sensor operating status. The operating state of the illustrated embodiment is an indication that the sensor is functioning correctly (eg, OK), as shown in data readings 1004m, 1004k, 1004o, and 1004s, respectively, and the sensor is in a power off state. An indication of something (eg, OFF), an indication that the sensor is reporting a single value (eg, STUCK), and / or an indication that the communication link to the network is down (eg, STUCK) For example, COM_DOWN) is included. In other embodiments, additional information and / or different information regarding the operational status of the traffic sensor may be provided, or such operational status information may not be available. Other traffic sensors, such as traffic sensors 123 and 166 in this example, are not configured to provide an indication of traffic sensor status, as indicated by "-" in traffic sensor status column 1002d.

  Rows 1004e, 1004j, 1004n, 1004q, 1004v, and 1004y and column 1002e can record additional traffic sensor data readings in some embodiments and / or provide additional information as part of each data reading. And / or can be recorded. Similarly, in some embodiments, less information than shown can be utilized as the basis for the techniques described herein.

  FIG. 10B shows an example of detection of an error in reading traffic sensor data that can indicate an unhealthy traffic sensor operating illegally. In particular, a large number of traffic sensors may not be able to provide an indication of traffic sensor status, and in some cases such traffic sensor status indications are unreliable (eg, in practice not Sometimes using statistical and / or other techniques, as it may indicate that the sensor is not functioning properly, or in fact it indicates that the sensor is functioning correctly when it is not) It may be desirable to detect unhealthy traffic sensors based on reported data readings.

  For example, in some embodiments, a plurality of current distributions of data readings reported by a given traffic sensor during a time period of a particular day (eg, between 4:00 PM and 7:29 PM) An unhealthy traffic sensor can be detected by comparing the historical distribution of data readings reported by that traffic sensor during the same time period over a past day (eg, the past 120 days). Such a distribution can be generated, for example, by processing multiple data reads obtained from a traffic sensor such as that shown in FIG. 10A.

  FIG. 10B shows three histograms 1020, 1030, and 1040, each representing a data read distribution based on data read obtained from the traffic sensor 123 during the time period of interest. The data represented by the histograms 1020, 1030, and 1040 is discretized at intervals of 8 km (5 miles) per hour (eg, 0 to 6.4 km per hour (0 to 4 miles), 8 to 14.4 km per hour (from 5 9 mile), 16 to 22.4 km per hour (10 to 14 mile, etc.), normalized, each bar (eg, bar 1024) is the vehicle speed in that bar 8 km (5 mph) bucket per hour It represents the probability between 0 and 1 that occurred during the period (eg, based on the percentage of data read during the time period contained in the bucket). For example, Bar 1024 reported that about 23% of the data readings obtained from traffic sensor 123 reported speeds between 80 km / h and 86.4 km / h (50 and 54 miles) including both ends. For example, the vehicle speed between 80 km / h and 86.4 km / h (50 and 54 miles) is observed by the traffic sensor 123 with a probability of about 0.23. In other embodiments, one or more bucket sizes can be used, either in addition to or instead of 8 km / h (5 mph) buckets. For example, a 1.6 km / h (1 mph) bucket can result in a finer granularity of processing, but also causes high variability between adjacent buckets when sufficient data reads are not available for the time period. While, the 16 km / h (10 mph) bucket provides lower variability but provides less detail. Further, although the current example uses average speed as a measure of data load analysis and comparison, in other embodiments, one or more other, either instead of or in addition to average speed. Can be used. For example, in at least some embodiments, traffic and / or occupancy can be used as well.

  In this example, histogram 1020 represents the historical distribution of data readings taken by traffic sensor 123 between 9:00 AM and 12:29 PM on Monday over the last 120 days. Histogram 1030 represents the distribution of data readings taken by sensor 123 between 9:00 AM and 12:29 PM on a particular Monday when traffic sensor 123 was functioning correctly. If the traffic pattern for a particular Monday is expected to resemble the traffic pattern for Monday in general, it can be visually identified that the shape of the histogram 1030 is similar to the histogram 1020, with a similar degree of Can be computed in various ways, as described below. Histogram 1040 shows that between 9:00 AM and 12:29 on a specific Monday when the traffic sensor 123 is not functioning properly and instead outputs a data read that does not reflect the actual traffic flow. It represents the distribution of data reading taken by the traffic sensor 123. The shape of the histogram 1040 is clearly different from the shape of the histogram 1020, as can be visually identified, and reflects the erroneous data reading reported by the traffic sensor 123. For example, a highly distributed spike is visible at bar 1048, which indicates that sensor 123 is stuck during at least part of the time between 9:00 AM and 12:30 PM and does not reflect actual traffic flow. May indicate that it was reporting a significant number of identical reads.

  In some embodiments, the Kullback-Leibler information amount between two traffic sensor data distributions can be utilized to determine the similarity between the two distributions, while in other embodiments, the distribution Similarity or difference between can be calculated in other ways. The Kullback-Leibler information amount is a convex measure of the similarity between the two probability distributions P and Q. This can be expressed as:

Where P _i and Q _i are the values of the discretized probability distributions P and Q (eg, each P _i and Q _i is the probability that the velocity occurred in the i-th bucket). In the illustrated example, the Kullback-Leibler information amount (“DKL”) 1036 between the data reading distribution shown in the histogram 1020 and the data reading distribution shown in the healthy traffic sensor histogram 1030 is about 0.076. However, the Kullback-Leibler information amount 1046 between the data reading distribution shown in the histogram 1020 and the data reading distribution shown in the unhealthy traffic sensor histogram 1040 is about 0.568. As expected, DKL 1036 is significantly smaller than DKL 1046 (in this case, approximately 13% of DKL 1046), which represents histogram 1030 (eg, the output of traffic sensor 123 while functioning correctly). ) Is more similar to the histogram 1020 (eg, representing the average behavior of the traffic sensor 123) than the histogram 1040 (eg, representing the traffic sensor 123 while malfunctioning) is more similar to the histogram 1020. It reflects.

  Further, some embodiments use other statistical measures, such as statistical information entropy, either instead of or in addition to similarity measures such as from the Kullback-Leibler information amount. It is possible to detect erroneous data readings supplied by traffic sensors. The statistical entropy of a probability distribution is a measure of the diversity of the probability distribution. The statistical entropy of the probability distribution P can be expressed as:

Here, P _i is a value of the discretized probability distribution P (for example, each P _i is a probability that the speed has occurred in the i-th bucket of the histogram of P). In the example shown, the statistical entropy 1022 of the distribution shown in histogram 1020 is about 2.17, and the statistical entropy 1032 of the distribution shown in histogram 1030 is about 2.14, The statistical entropy 1042 for the distribution shown is about 2.22. As expected, the statistical entropy 1042 is larger than both the statistical entropy 1032 and the statistical entropy 1022, which represents the more chaotic output pattern exhibited by the traffic sensor 123 while malfunctioning. It reflects.

Furthermore, the difference between two statistical entropy measurements can be measured by calculating an entropy difference measurement. The measured entropy difference between the two probability distributions P and Q is
EM = ‖H (P) −H (Q) ‖ ²
Where H (P) and H (Q) are the entropies of the probability distributions P and Q described above, respectively. In the illustrated example, the entropy difference measurement (“EM”) 1034 between the distribution shown in histogram 1020 and the distribution shown in histogram 1030 is about 0.0010, and the distribution shown in histogram 1020 is shown. And the measured entropy difference 1044 between the distribution shown in the histogram 1040 is about 0.0023. As expected, the entropy difference measurement 1044 is significantly larger (more than twice as large in this case) as the entropy difference measurement 1034, which is the statistical entropy of the distribution shown in the histogram 1030 and the histogram 1020. Reflects the greater difference between the statistical entropy of the distribution shown in histogram 1040 and the statistical entropy of the distribution shown in histogram 1020 compared to the difference between the statistical entropy of the distribution shown in FIG. To do.

  The statistical measurements described above can be used in various ways to detect unhealthy traffic sensors. In some embodiments, various information about the current data load distribution is provided as input to sensor health (or data load reliability) classifiers such as those based on neural networks, Bayesian classifiers, decision trees, etc. Is done. For example, classifier input information includes, for example, the amount of Kullback-Leibler information between the historical data reading distribution of a traffic sensor and the current data reading distribution of that traffic sensor, and the statistical entropy of the current data reading distribution. be able to. The classifier then evaluates the health of the traffic sensor based on the supplied input and provides an output indicating an unhealthy sensor or a healthy sensor. In some cases, the time corresponding to the current data reading distribution and the historical data reading distribution and / or the time corresponding to the day of the week (eg, the time period from 5:00 AM to 9:00 AM), the day, or the day of the week (eg, , Monday to Thursday, Friday, Saturday, or Sunday), additional information such as the size of the mph bucket can also be supplied as input to the classifier. The classifier can be trained by utilizing actual previous data readings, such as those including the traffic sensor status indication shown in FIG. 10A.

  In other embodiments, unhealthy traffic sensors can be identified without using a classifier. For example, a traffic sensor can be determined to be unhealthy when one or more statistical measurements exceed a predetermined threshold. For example, if the amount of Kullback-Leibler information between the historical data reading distribution of a traffic sensor and the current data reading distribution of the traffic sensor exceeds a first threshold, the statistical entropy of the current data reading distribution is a second threshold. And / or if the entropy difference between the current data reading distribution and the historical data reading distribution exceeds a third threshold, the traffic sensor may be determined to be unhealthy. In addition, other non-statistical information can be utilized, such as whether the traffic sensor reports a sensor condition that can be interpreted as healthy or unhealthy.

  As noted above, the above technique is described primarily in the context of traffic sensors reporting vehicle speed information, but the same technique applies to other traffic including traffic, traffic density, and traffic occupancy. It can be used for flow information.

  FIG. 3 is a block diagram illustrating an embodiment of a computing system 300 suitable for performing at least some of the described techniques, such as by performing an embodiment of a data sample manager system. The computing system 300 includes a central processing unit (“CPU”) 335, various input / output (“I / O”) components 305, storage 340, and memory 345, the illustrated I / O components include a display 310, Includes network connection 315, computer readable media drive 320, and other I / O devices 330 (eg, keyboard, mouse or other pointing device, microphone, speaker, etc.).

  In the illustrated embodiment, data sample manager system 350, predictive traffic information provider system 360, major road identifier system 361, road segment determiner system 362, RT information provider system 363, and other options provided by program 369. A variety of systems are running in memory 345 to implement at least some of the described techniques, and these various running systems are generally referred to herein as traffic information systems. Called. Computing system 300 and its running systems include various client devices 382, vehicle-based clients and / or data sources 384, road traffic sensors 386, other data sources 388, and others such as third party computing systems 390. Can be communicated with a computing system over a network 380 (eg, the Internet, one or more cellular telephone networks, etc.).

  In particular, the data sample manager system 350 may provide current traffic conditions from various sources, such as from road traffic sensors 386, vehicle-based mobile data sources 384, and / or other mobile or non-mobile data sources 388. Or obtain a variety of information about previously observed case data. The data sample manager system 350 then filters the obtained data to other data by filtering the data (eg, excluding data samples from consideration) and / or conditioning the data (eg, correcting errors). Prepare for use by components and / or systems, and then use the prepared data to evaluate road traffic conditions such as traffic flow and / or speed for various road segments. In the illustrated embodiment, the data sample manager system 350 includes a data sample fitter component 352, a sensor data conditioner component 353, a data sample outlier eliminator component 354, a data sample rate assessor component 356, a data sample flow assessor component 358, and 1 includes optional sensor data aggregator components 355 that correspond to the corresponding components of FIG. 1 (data sample filter component 104, sensor data conditioner component 105, data sample outlier eliminator component 106, data sample rate assessor component 107, data sample flow assessor computer Nento 108, and the like optional sensor data aggregator component 110) to perform a function similar to the function described above for. Further, in at least some embodiments, the data sample manager system may be substantially Perform road traffic assessment in real time or near real time.

  Thereafter, other traffic information systems 360-363 and 369 and / or third party computing system 390 may use the data provided by the data sample manager system in various ways. For example, the predictive traffic information provider system 360 obtains such prepared data (directly or indirectly through a database or storage device) to generate future traffic forecasts at multiple future times. One or more other traffic information systems, client devices 382, vehicle-based clients 384, and / or third party computing systems 390, etc. to one or more other recipients Predictive information can be supplied. In addition, the RT information provider system 363 obtains information about the evaluated road traffic situation from the data sample manager system and obtains the road traffic situation information in real-time or near real-time form to others (eg, client device 382, vehicle-based). Client 384, and / or third party computing system 390), and the data sample manager system can also be used from the RT information provider system when performing its evaluation in such real-time or near real-time form. The data recipients of this data may receive concurrent actual vehicle movement on one or more road segments (by mobile data sources traveling on these road segments and / or these road segments). May be able to see and use information about current traffic conditions on these road segments based on sensors and other data sources that provide information about actual vehicle movement on the road) is there.

  The client device 382 can take various forms in various embodiments and is generally any communication device and other that can make requests to and / or receive information from the traffic information system. A computing device can be included. In some cases, the client device is an interactive console that a user can use to request traffic related information (eg, predicted future traffic status information, real-time or near real-time current traffic status information, etc.) An application (eg, a web browser) can be run, but in other cases, at least some such traffic related information is automatically sent to the client device from one or more of the traffic information systems (E.g., as a text message, as a new web page, as a specialized program data update, etc.).

  The road traffic sensor 386 includes a plurality of sensors installed in or near various streets, highways, or other roads, such as for one or more geographic areas. These sensors may include loop sensors that can measure the number of vehicles passing over that sensor per unit time, vehicle speed, and / or other data related to traffic flow. In addition, such sensors include cameras, motion sensors, radar distance measuring devices, RFID-based devices, and other types of sensors that are located close to or otherwise close to the road Can do. The road traffic sensor 386 uses one or more data exchange mechanisms (eg, push, pull, poll, request-response, peer-to-peer, etc.) to wire-based or wireless to the data sample manager system 350 over the network 380. Through the base data link, measured data readings can be provided periodically or continuously. Further, although not shown herein, in some embodiments, one or more aggregators of such road traffic sensor information (eg, a government transport that operates the sensor) may instead Data can be obtained and made available from the traffic information system (either in raw form or after it has been processed).

  Other data sources 388 are other sources of various types of data that can be utilized by one or more of the traffic information systems to provide traffic related information to users, customers, and / or other computing systems. including. Such data sources include road networks such as the connectivity of various roads to each other and the traffic control information associated with such roads (eg, the presence and / or location of traffic control signals and / or speed limit segments). A map service and / or map database that provides information about Other data sources include short and long term weather forecasts, school schedules and / or calendars, event schedules and / or calendars, human operators (eg, first responders, law enforcement officers, highway crews, news Including sources of information about events and / or conditions that affect and / or reflect traffic conditions, such as traffic accident reports, road construction information, vacation schedules, etc. provided by media, travelers, etc.) it can.

  The vehicle-based client / data source 384 in this example each provides data to one or more of the traffic information systems and / or receives data from one or more of the traffic information systems. It may be a computing system and / or a communication system located within. In some embodiments, the data sample manager system 350 may provide vehicle-based mobile data sources and / or other user-based mobile data sources that provide information related to current traffic conditions for use by the traffic information system. A distributed network (not shown) can be used. For example, each vehicle or other mobile data source can determine the geographic location associated with the movement of the vehicle, and possibly other information such as speed, direction, elevation, and / or other data (“global positioning” System ") devices (e.g., cell phones with GPS capabilities, stand-alone GPS devices, etc.) and / or other geolocation devices, such as geolocation device (s) or other separate communications The device sometimes obtains such data and provides it to one or more of the traffic information systems (eg, via a wireless link).

  Alternatively, some or all of the vehicle-based client / data source 384 may be in-vehicle to obtain information from one or more of the traffic information systems, such as for use by vehicle occupants. Can have a computing system and / or a communication system. For example, an installed vehicle that can be used by a user to request traffic related information over a wireless link from one of the traffic information systems, such as a predictive traffic information provider system and / or an RT information provider system. An in-dashboard navigation system with a web browser or other console application can be included, or alternatively, such a request can be made from a user's portable device in the vehicle. Further, one or more of the traffic information systems can automatically transmit traffic related information to such vehicle-based client devices based on receipt or generation of updated information.

  Third party computing system 390 may be other than a traffic information system operator (s), such as a party that receives traffic related data from one or more of the traffic information systems and a party that uses the data in some way. Includes one or more optional computing systems operated by a party. For example, the third party computing system 390 may receive traffic information from one or more of the traffic information systems and provide relevant information (either received information or other information based on the received information). It can be a system that supplies to users or others (eg, via a web portal or subscription service). Alternatively, the third-party computing system 390 can be used for other types of media such as online map companies that collect traffic conditions and report traffic to consumers as part of a media organization or travel planning service. Can be operated by a party.

  As previously noted, the predictive traffic information provider system 360 was prepared by the data sample manager system 350 and other components in the illustrated embodiment to generate future traffic forecasts for multiple future times. Data can be used. In some embodiments, the prediction repeats a future time series prediction of each of a number of road segments, such as in real time based on changing current conditions for a network of roads within a given geographic area. To create, it is generated using probabilistic techniques that incorporate various types of input data. Further, in at least some embodiments, one or more predictive Bayes models or other models (eg, decision trees) are based on observed historical traffic conditions of the geographic area of interest, etc. Created automatically for use in generating future traffic forecasts for each of these geographic areas. Predicted future traffic information can be used in various ways to assist in travel and for other purposes, such as planning the optimal route through a network of roads based on predictions about road traffic conditions at multiple future times. Can be used.

  In addition, the road segment determiner system 362 can be used by other traffic information systems to automatically determine and manage road-related information within one or more geographic areas. A map service and / or a map database that provides information about can be used. Such road-related information includes the determination of specific parts of the road that should be treated as a road segment of interest (for example, based on traffic conditions on these road parts and other nearby road parts), An automatically generated association or relationship between a road segment in a given road network and the display of other information of interest (eg, physical location of road traffic sensors, event locations, and Landmarks, functional road types and other relevant traffic characteristics, etc.). In some embodiments, the road segment determiner system 362 runs periodically and stores the information it creates in a storage 340 or database (not shown) for use by other traffic information systems. be able to.

  In addition, the main road identifier system 361 automatically identifies the roads that are of interest for tracking and evaluating road traffic conditions, such as for use by other traffic information systems and / or traffic data clients. To identify, the road network representing a given geographic area and the traffic status information for that geographic area are used. In some embodiments, automatic identification of a road (or one or more road segments of the road) as being of interest may include peak traffic volume or other flow magnitudes, peak traffic congestion Intra-day variability of size, traffic or other flows, intra-day variability of road congestion, inter-day variability of traffic or other flows, and / or It can be based at least in part on factors such as daily variability of road congestion. Such factors may be, for example, first calculating the covariance matrix S of traffic conditions information for all roads (or road segments) within a given geographic area, and then Eigen decomposition of the covariance matrix S. ) To calculate by principal component analysis. Next, in descending order of eigenvalues, the eigenvectors of S represent the combinations of roads (or road segments) that most strongly and independently contribute to the observed traffic situation variance.

  Further, the real-time traffic information provider system or the real-time traffic information presenter system can be provided by one or more of the other programs 369 by the RT information provider system or alternatively. The information provider system may provide client device 382, vehicle-based client 384, third party, such as to provide data in real-time or near real-time form based at least in part on data samples obtained from vehicles and other mobile data sources. Data sample manager system 350 and / or other components (predicted traffic information) to provide traffic information services to consumers and / or business entities operating or otherwise utilizing party computing system 390 or the like. Data that is analyzed and supplied by a provider system 360 or the like can be used.

  It will be appreciated that the computing system shown is exemplary only and is not intended to limit the scope of the invention. The computing system 300 may be connected to other devices not shown, including via one or more networks such as the Internet or via the web. More generally, a “client” or “server” computing system or computing device or traffic information system and / or component is a desktop computer or other computer, database server, network storage server and other network device, PDA, Cell phones, wireless phones, pagers, electronic organizers, Internet appliances, television-based systems (eg, using set-top boxes and / or personal / digital video recorders), and various others including appropriate intercommunication functions Hardware that can interact and perform the types of functionality described, including, without limitation, It is possible to include all of the combinations of software. Further, the functionality provided by the illustrated system components can be combined into fewer components or distributed over additional components in some embodiments. Similarly, in some embodiments, some functionality of the illustrated components may not be provided and / or other additional functionality may be enabled.

  Further, although various items are illustrated as being stored in memory or on storage while in use, these items or portions thereof may be used for memory management and / or data integrity. It can be transferred between memory and other storage devices. Instead, in other embodiments, some or all of the software components and / or modules can execute in the memory of another device and communicate with the illustrated computer system via computer-to-computer communication. Some or all of the system components or data structures may also be stored on a computer readable medium such as a hard disk, memory, network, or portable media article that is read by a suitable device or via a suitable connection (eg, software As instruction or structured data). System components and data structures are generated on various computer readable transmission media including wireless and wired / cable based media (eg, carrier waves or portions of other analog or digital propagated signals) System components and data structures can take a variety of forms (eg, as part of a single or multiplexed analog signal, or as multiple discrete digital packets or discrete digital As a frame). Such computer program products may take other forms in other embodiments. Thus, the present invention can be practiced using other computer system configurations.

  FIG. 4 is a flowchart of an example embodiment of a data sample filter routine 400. This routine may, for example, receive data samples corresponding to roads in the geographic area and filter data samples that are not of interest for later evaluation, such as the data sample phytara component 352 and / or FIG. It can be implemented by execution of the embodiment of the data sample filter component 104 of FIG. The filtered data sample is then used to calculate the average speed of a particular road segment of interest and to calculate the filtered data sample to calculate other traffic flow related characteristics of such road segment. It can be used in various forms, such as for use.

  The routine begins at step 405 where a group of data samples for a particular period of time for a geographic area is received. In step 410, the routine then optionally generates additional information about some or all of the data samples based on other related data samples. For example, if a particular data sample of a vehicle or other mobile data source lacks information of interest (such as the speed and / or direction or direction of travel of the mobile data source), such information , And can be determined in relation to one or both of previous and subsequent data samples of the same mobile data source. Further, in at least some embodiments, to evaluate data source activity over a period of time across multiple data samples (eg, as part of the normal flow of traffic, such as a vehicle with a stop sign or stop signal) Multiple data samples for a particular mobile data source to evaluate additional types of information about the data source, such as to determine if you have been parked for several minutes instead of temporarily stopping for ~ 2 minutes) Information from can be aggregated.

  After step 410, the routine continues to step 415 and attempts to associate each data sample with a road in the geographic area and a particular road segment of the road, but in other embodiments, at least the road and / or road. If the initial association of data samples to a segment is received instead at step 450, or instead this entire routine is performed at a single time for a single road segment and all of the data samples received as a group at step 405 This step may not be performed, or may be performed in other ways, such as when corresponds to a single road segment. In the illustrated embodiment, the association of data samples to roads and road segments is for an initial association based solely on the geographic location associated with the data samples (eg, to associate data samples with the closest road and road segment). It can be executed in various forms. In addition, the association can optionally include additional analysis to refine or revise its initial association, for example, location-based analysis can include multiple possible road segments for a data sample (multiple of specific roads). Such as multiple road segments of roads that are close or otherwise unrelated to other road segments), such additional analysis may include other information such as speed and heading to influence the association Can be used (e.g., by combining location information and one or more other such factors in a weighted manner). Thus, for example, if the reported location of a data sample is between the freeway and a nearby side road, information about the reported speed of the data sample is used to associate the data sample with the appropriate road. (E.g., by determining that a data sample with an associated speed of 112 km / h (70 miles) is unlikely to originate from a side road with a speed limit of 40 km / h (25 miles)) . Further, a particular stretch of road or other road portion is associated with a plurality of separate road segments (eg, movement in one direction is modeled as a first road segment and movement in the opposite direction is a separate second road. Two-lane roads modeled as segments, or alternatively multi-lane freeways where HOV lanes are modeled as separate road segments from one or more adjacent non-HOV lanes), such as speed and / or direction Additional information about the data sample can be used to select the most likely road segment for that road for the data sample.

  After step 415, the routine continues to step 420 where it is not associated with a road segment that is of interest for later processing, including data samples (if any) that are not associated with any road segment. Filter all data samples. For example, to exclude roads of a certain functional road type (eg if the road size and / or the amount of traffic is not large enough to be of interest) or freeway ramps or branch roads or collectors / distributors Certain roads or portions of roads may not be of interest for later analysis, such as to exclude such road portions because the traffic characteristics of the portions of the road do not represent the overall freeway. Similarly, to exclude freeway HOV lanes in situations where multiple road segments are associated with a particular part of the road and only the behavior of non-HOV lanes is of interest for a particular purpose or 2 Some road segments may be uninterested for a purpose, such as when only one direction of a directional road is interested. After step 420, the routine continues to step 425 to determine whether to filter the data sample based on data source activity, but in other embodiments, such filtering should not be performed, Must be executed. In the illustrated embodiment, if filtering is performed based on source activity, the routine continues to step 430 and corresponds to a data source whose behavior does not reflect the traffic flow activity of interest being measured. Data samples to be removed (for example, to exclude vehicles parked with the engine running for an extended period of time, in a parking lot, in a parking garage, or for an extended period of time, or Perform such filtering, such as to exclude vehicles that are driving in other narrow areas). If step 425 determines that no filtering based on data source activity is performed after or instead of step 430, the routine continues to step 490, where the filtered data is stored for later use. In other embodiments, the filtered data can instead be provided directly to one or more clients. The routine then continues to step 495 and determines whether to continue. If so, the routine returns to step 405, otherwise it continues to step 499 and ends.

  FIG. 5 is a flow diagram of an example embodiment of a data sample outlier eliminator routine 500. This routine may include, for example, the data sample outlier eliminator component 354 of FIG. 3 and / or the data sample outlier eliminator of FIG. 1, such as to remove road segment data samples that are outlier with respect to other data samples of the road segment. It can be realized by executing the embodiment of the component 106.

  The routine begins at step 505 where a set of data samples for a road segment and time period is received. The received data sample can be, for example, a filtered data sample obtained from the output of the data sample filter routine. Next, at step 510, the routine optionally separates the data samples into groups to reflect distinct portions and / or distinct behaviors of the road segment. For example, traffic flow during a time period when multiple freeway lanes are included together as part of a single road segment and the multiple lanes include at least one HOV lane and one or more non-HOV lanes. Vehicles in the HOV lane (s) can be separated from vehicles in the other lanes if are significantly different between HOV lanes and non-HOV lanes. Such groupings can come in a variety of ways, such as by fitting the data samples to multiple curves, each representing a normal data sample variability within a particular group of data samples (eg, a normal curve or a Gaussian curve). Can be executed. In other embodiments, if a road segment is instead split such that all data samples for that road segment reflect similar behavior (eg, freeways with HOV lanes and other non-HOV lanes) Such as when it is divided into a plurality of road segments instead).

  The routine then continues to step 515 for each group or groups of data samples (if the data sample separation of step 510 is not performed, all of the data samples are treated as a single group) Calculate all average traffic characteristics of the data samples. Such average traffic situation characteristics may include, for example, corresponding statistical information such as average speed as well as standard deviation from the average value. The routine then continues to step 520 for each one or more data sample groups, selecting specific target data samples to tentatively ignore, and average traffic conditions for the remaining traffic condition characteristics. Only one leave-one-out analysis is performed in succession to determine the characteristics. The greater the difference between the average traffic condition characteristics of the remaining data samples and the average traffic condition characteristics of all data samples from step 515, the more the excluded target data sample has the common characteristics of the other remaining data samples The likelihood of an abnormal value that is not reflected is high. Next, at step 525, the routine optionally performs one or more additional types of outlier analysis, such as to sequentially exclude groups of multiple target data samples to assess joint effects. However, in some embodiments, such additional outlier analysis may not be performed. After step 522, the routine continues to step 590, removing the data samples identified as outliers in steps 520 and / or 525 and storing the remaining data samples for later use. In other embodiments, the routine can instead transfer the remaining data samples to one or more clients for use. The routine then continues to step 595 to determine if it should continue. If so, the routine returns to step 505; otherwise, the routine continues to step 599 and ends.

  FIG. 6 is a flow diagram of an example embodiment of a data sample rate assessor routine 600. This routine may include, for example, the data sample speed assessor component 356 of FIG. 3 and / or the data sample of FIG. 1, such as to evaluate the current average speed of the road segment over a period of time based on various data samples of the road segment. This can be realized by executing the speed assessor component 107. In this example embodiment, this routine performs a continuous calculation of the average speed of the road segment during each of a plurality of time intervals or time windows over a period of time, while in other embodiments, each of the routines The call may instead be for a single time interval (eg, multiple time intervals are evaluated via multiple calls of this routine). For example, if the time period is 30 minutes, a new average speed calculation can be made using a 5 minute time interval (thus using each time interval that does not overlap with the preceding and following time intervals) or a 10 minute time interval. Can be run every 5 minutes, such as with (and thus overlapping with adjacent time intervals).

  The routine begins at step 605 where a period of road segment data samples (eg, data samples from mobile data sources and physical sensor data reads) or a period of road segment insufficiency. However, in some embodiments, only one of the data sample from the mobile data source and the data sample from the sensor data reading may be received. The received data sample can be obtained, for example, from the output of the data sample outlier eliminator routine. Similarly, an indication of insufficient data can be received from the data sample outlier eliminator routine. In some cases, the display of inadequate data may be due to a lack of data samples from mobile data sources associated with the road segment for a period of time and / or some or all sensor data readings for the road segment May be based on having an insufficient number of data samples, such as when it is detected that there is a missing or error (eg, by sensor data conditioner component 105 of FIG. 1). In this example, the routine continues at step 610 to determine whether an indication of insufficient data has been received. If so, the routine continues to step 615; otherwise, the routine continues to step 625.

  At step 615, the routine executes an embodiment of a traffic flow estimator routine (described with respect to FIG. 14) to obtain an estimated average traffic speed for the road segment over a period of time. Next, at step 620, the routine provides an indication of the estimated average speed. In step 625, the routine selects the next time interval or time window where the average rate must be evaluated, starting with the first time interval. Next, at step 630, the routine calculates a weighted average traffic speed for the data samples within the time interval, and the weighting of the data samples is based on one or more factors. For example, in the illustrated embodiment, the weighting of each data sample gives greater weight to data samples near the end of the time interval (which can more strongly reflect the actual average rate at the end of the time interval). (Eg, linearly, exponentially, or in a step-wise manner). Further, in the illustrated embodiment, data samples, such as to weight data readings from physical sensors, whether more or less, differently from data samples from vehicles and other mobile data sources, etc. Can be weighted based on the source of the data. Furthermore, in other embodiments, various other factors can be used for weighting, including the basis for each sample, for example, available information about the sensor (eg, one of the physical sensors is intermittent). Data reading from one physical sensor is different from data reading from another physical sensor, such as to reflect failure of the system or to have a data reading resolution that is less accurate than another sensor) Data samples from one vehicle or other mobile data source, as well as data samples from another such vehicle or other mobile data source based on information about that mobile data source Can be weighted differently. Other types of factors that can be used for weighting in some embodiments are confidence values for specific data samples or other estimates of possible errors, that a specific data sample must be associated with a specific road segment Including the degree of reliability.

  In step 630, the routine continues to step 635 and provides an indication of the calculated average traffic speed for the time interval, such as to store information for later use and / or to supply information to the client. To do. Next, at step 640, the routine optionally obtains additional data samples for the time period that became available after receipt of the information at step 605. In step 645, it is determined whether additional time intervals should be calculated for the time period, and if so, the routine returns to step 625. Instead, if there are no further time intervals, or after step 620, the routine continues to step 695 to determine whether to continue. If so, the routine returns to step 605, otherwise it continues to step 699 and ends.

  FIG. 7 is a flow diagram of an example embodiment of a data sample flow assessor routine 700. This routine may be used to evaluate traffic flow characteristics other than the average speed of a particular road segment during a particular period of time, such as the data sample flow assessor component 358 of FIG. 3 and / or the data sample flow assessor component 108 of FIG. This can be realized by executing the embodiment. In this example embodiment, the flow characteristics that are evaluated include the total amount of vehicles (or other mobile data sources) that arrive at or reside on a particular road segment over a period of time and points or road segments on the road segment. The percentage occupancy of the road segment during a certain time period reflects the percentage of time that the area is covered by the vehicle.

  The routine begins at step 705, where a data sample of a road segment over a period of time and an average speed of the road segment over the period of time or an indication of insufficient data of the road segment over the period of time is received. . The data sample can be obtained, for example, from the output of the data sample outlier eliminator routine, and the average rate can be obtained, for example, from the output of the data sample rate assessor routine. An indication of insufficient data can be obtained, for example, from the output of the data sample outlier eliminator routine. In some cases, the display of insufficient data is missing when there is no data sample from a mobile data source associated with a road segment for a period of time and / or some or all sensor data readings for the road segment Or based on having an insufficient number of data samples, such as when it is detected that there is an error (eg, by the sensor data conditioner component 105 of FIG. 1). The routine then continues to step 706 to determine whether an indication of insufficient data has been received. If so, the routine continues to step 750; otherwise, the routine continues to step 710.

  In step 750, the routine executes an embodiment of a traffic flow estimator routine (described with respect to FIG. 14) to obtain an estimated total amount and occupancy of road segments over a period of time. Next, at step 755, the routine provides an indication of the estimated total amount and occupancy.

  In step 710, the routine determines the number of vehicles (or other mobile data sources) that supply the data samples, such as by associating each data sample with a particular mobile data source. Next, at step 720, the routine probabilistically determines the most likely arrival rate to the road segment of the vehicle supplying the data samples based in part on the determined number of vehicles. In some embodiments, the probabilistic determination may further use information about the a priori probability of the number of such vehicles and the a priori probability of a particular arrival rate. Next, at step 730, the routine will run through all road segments during a period of time, such as based on information about the determined number of vehicles and which percentage of the total number of vehicles is the vehicle supplying the data samples. The total amount of vehicles is inferred, and the confidence interval of the inferred total amount is evaluated. Next, at step 740, the routine infers the percentage occupancy of the road segment over a period of time based on the inferred total amount, average speed, and average vehicle length. In other embodiments, other types of traffic flow characteristics of interest can be evaluated as well. In the illustrated embodiment, the routine then continues to step 790 to provide an indication of the inferred total amount and the inferred percentage occupancy. If after step 755 or 790 it is determined to continue at step 795, the routine returns to step 705, otherwise it continues to step 799 and ends.

  FIG. 11 is a flowchart of an example embodiment of a sensor data read error detector routine 1100. This routine may be implemented by execution of the sensor data conditioner component 353 of FIG. 3 and / or the sensor data conditioner component 105 of FIG. 1, for example, to determine the health of one or more traffic sensors. it can. In this example embodiment, this routine is at various times to determine the health of one or more traffic sensors based on traffic sensor data readings recently obtained during the indicated time period. Executed. In addition, data output by the traffic sensor for one or more of various types of traffic condition measurements, such as traffic speed, traffic volume, traffic occupancy, etc., can be analyzed by this routine in various embodiments. . Further, at least some data of traffic conditions is measured and / or aggregated in various forms, such as at various levels of granularity (eg, 8 km / h (5 mph) buckets of a group of speed information data). This routine may, in some embodiments, be specific traffic sensors at each of one or more levels of granularity (or other levels of aggregation) for each of one or more traffic condition measurements. Can be analyzed.

  This routine begins at step 1105 with one or more traffic sensors and a selected time category (eg, if this routine is run after each time category to provide results in near real-time form. Receive the display of the most recent time category or one or more previous time categories selected for analysis, but in other embodiments, multiple time categories can be shown instead. . In some embodiments, each time is a time category (eg, 12:00 AM to 5:29 AM and 7:30 PM to 11:59 PM, 5:30 AM to 8:59 AM, 9:00 AM to 12: Up to 29PM, 12:30 PM to 3:59 PM, 4:00 PM to 7:29 PM, and 12:00 AM to 11:59 PM) and / or day of week category (eg, Monday through Thursday, Friday, Saturday, and Sunday) , Or alternatively group Saturday and Sunday together) with each containing time category. A particular time category may, in various embodiments, have similar characteristics of traffic, such as to group these times when traffic is relatively sparse during normal evening and early morning hours. It can be selected in a variety of ways, including to reflect the time period between expectations (eg, based on commuting time and commuting patterns or other consistent activities that affect traffic). Further, in some embodiments, the time categories may be different traffic sensors, either manually or in an automatic manner by analyzing historical data to determine time periods having similar traffic flow characteristics. Can be selected differently (eg, by geographical area, road, individual sensors, etc.).

  Next, in steps 1110 through 1150, the routine executes a loop, in which the time category indication is used to determine the traffic sensor health status of each of the traffic sensors in the indicated time category. Analyzing the traffic sensor data reading from each of the one or more of the traffic sensors. In step 1110, the routine starts with the first traffic sensor, selects the next traffic sensor of the indicated one or more traffic sensors, and in the indicated time category (or alternatively, multiple in step 1105). If a time category is indicated, select the traffic sensor and the next combination of indicated time categories). In step 1115, the routine retrieves the average historical data reading distribution of traffic sensors during the selected time category. In some embodiments, the historical data read distribution is in a selected time category over an extended time period, such as the last 120 days or the last 120 days (eg, 4 days of the week, including Monday through Thursday). Between 0:00 PM and 7:29 PM) based on data readings supplied by the traffic sensor).

  In step 1120, the routine determines the target traffic sensor data distribution for the selected traffic sensor and the selected time category. Next, in step 1125, the routine determines the similarity between the target traffic sensor data reading distribution and the historical traffic sensor data reading distribution. As described in more detail elsewhere, in some embodiments, such similarity measures are used to calculate the amount of Kullback-Leibler information between the target traffic sensor data reading distribution and the historical traffic sensor data reading distribution. It can be determined by doing. Next, at step 1130, the routine determines the information entropy of the target traffic sensor data reading distribution, as described in more detail elsewhere.

  Next, at step 1135, the routine uses various information to indicate a health classification (eg, a “healthy” or “unhealthy” indication or a value on a “health” scale, such as 1 to 100). To evaluate the health of the selected traffic sensor in the selected time category, which in this example is the determined similarity, determined entropy, and selected time Categories (eg, selected time categories such as 4:00 PM to 7:29 PM and / or selected day categories such as Monday to Thursday). In other embodiments, other types of information may be used, such as an indication of the degree of granularity of the data being measured (eg, 8 km / h (5 mph) buckets for a group of speed information data). In one embodiment, a neural network can be used for classification, but in other embodiments, various other classification techniques can be utilized including decision trees, Bayesian classifiers, and the like.

  Next, at step 1140, the routine determines the traffic sensor health status of the selected traffic sensor and the selected time category (in this example, based on the evaluated traffic sensor health and / or other factors. As healthy or unhealthy). In some embodiments, the health status of the traffic sensor can be determined to be healthy whenever the traffic sensor health of the selected time category is evaluated as healthy at step 1135. Further, the health status of the traffic sensor is evaluated as unhealthy for the selected time category of traffic sensor (eg, at step 1135), and the selected time category is sufficiently long (eg, at least 12). It can be determined that it is unhealthy whenever it has an associated time category that covers time or 24 hours). Further, in some embodiments, information regarding the associated time category (eg, one or more previous and / or subsequent time periods) is obtained from traffic sensor health over a longer period of time (eg, one day). Can be taken out and used, such as for sorting. Such logic can detect false negatives of sensor health status based on temporary abnormal traffic patterns accurately reported by the traffic sensor (eg, traffic sensor health status when the traffic sensor is actually healthy) Can be reduced.

  For example, a false negative determination can occur due to substantial daytime variability in data reading due to external factors (eg, traffic accidents, weather incidents, etc.). For example, a car accident that occurs at or near a particular traffic sensor may result in that traffic sensor supplying abnormally incorrect data readings over a relatively short period of time (eg, 1 to 2 hours). If the sensor health status determination is based solely on data readings obtained primarily during times of disturbance caused by traffic accidents, false negative determinations are likely to occur. By making the determination of unhealthy sensor status based on data readings obtained over a relatively longer time period (eg, 12 hours or 24 hours), the risk of such false negative determinations can be reduced. . On the other hand, false-positive determination (for example, determining that the health of the traffic sensor is healthy when it is actually unhealthy) is that historical data is read by the malfunctioning traffic sensor (for example, reflecting normal traffic patterns) In general, it may be less likely because it is less likely to provide a data read similar to Therefore, it may be appropriate to determine the health status of the traffic sensor as healthy based on a relatively shorter time period.

  Some embodiments perform this routine multiple times daily using a time category that reflects a shorter time period (eg, using a time category with a time category that extends over the previous 3 hours, Run this routine at least once daily with a time category that reflects the entire previous day (e.g., this time at midnight with a time category that has a time category that extends over the previous 24 hours) Routines) can provide such discriminatory logic.

  Further, the determination of sensor health status can determine whether a sufficient number of data reads are available for the selected time category (eg, because the traffic sensor is reporting data reads intermittently) and / or the traffic sensor. May be based on other factors, such as based on an indication of the sensor status supplied by (e.g., that the traffic sensor is stuck).

  At step 1145, the routine provides the determined traffic sensor health status. In some embodiments, the traffic sensor health status is stored (eg, in a database or file system) for later use by other components (eg, sensor data aggregator component 110 of FIG. 1), and / or Or it can be fed directly to another component (eg, a data sample outlier eliminator component). At step 1150, the routine determines whether there are more traffic sensors (or a combination of traffic sensors and time categories) to process. If so, the routine continues to step 1110, otherwise it continues to step 1155 and performs other actions as appropriate. Such other actions include, for example, a historical data reading distribution (eg, for the last 120 days) for each of one or more time categories for each of a plurality of traffic sensors (eg, 1 daily). Recalculation), once a week, etc. By periodically recalculating the historical data reading distribution, this routine ensures accurate traffic sensor health without regard to gradually changing traffic conditions (eg, due to the start or completion of road construction projects). You can continue to provide situation determination. After step 1155, the routine continues to step 1199 and returns.

  FIG. 12 is a flowchart of an example embodiment of a sensor data read error collector routine 1200. This routine may include the sensor data conditioner component 353 of FIG. 3 and / or the sensor data conditioner component 105 of FIG. 1, such as to determine a corrected data read of one or more traffic sensors associated with a road segment. It can be realized by execution. In the illustrated example embodiment, this routine may be run periodically (eg, every 5 minutes) to correct traffic sensor data reads identified as unhealthy by the sensor data read error collector routine. . In other embodiments, this routine can be performed on demand, such as by a sensor data aggregator routine to obtain a corrected data reading for a particular road segment, or alternatively in various situations. This routine may not be used. For example, data analysis and correction can include multiple data samples such as traffic sensors and multiple types of data sources that can include one or more distinct types of mobile data sources (eg, traffic sensors). Determine whether (from the data source) provides sufficient data to analyze the traffic flow conditions for that particular road segment, and if so, do not perform correction of data from individual traffic sensors Thus, it can be executed more generally.

  The routine begins at step 1205, where a display of a road segment to which one or more traffic sensors are associated (eg, sensor data in which one or more of the associated traffic sensors have been classified as unhealthy). An indication of one or more time categories to be processed as an option (eg, at least one of the associated traffic sensors in between is classified as at least potentially unhealthy) and based on the results from the read error detector routine Time category). In other embodiments, the one or more traffic sensors that are of interest may be shown in other forms, such as by directly receiving an indication of the one or more traffic sensors. In steps 1210 through 1235, the routine executes a loop in which unhealthy traffic sensors in the indicated road segment are processed to show one or more time categories (eg, as shown in step 1205). Determine and supply corrected data readings for these traffic sensors in the time category).

  In step 1210, the routine starts with the first unhealthy traffic sensor and selects the next unhealthy traffic sensor in the indicated road segment. This routine includes one or more time categories indicated in step 1205, such as by selecting one of one or more time categories during which the traffic sensor was previously designated unhealthy. Also select a time category to use, such as one of them. At step 1215, the routine is healthy and whether there are enough other traffic sensors in the indicated road segment that can be used to assist in correcting the unhealthy traffic sensor readings for the selected time category. Determine. This determination is whether there are at least a predetermined number (eg, at least two) and / or a predetermined percentage (eg, at least 30%) of healthy traffic sensors in the road segment indicated in the selected time category. In addition, the relative position of healthy traffic sensors within the indicated road segment can be taken into account (e.g., adjacent or otherwise nearby traffic sensors, unhealthy Can be preferred over traffic sensors farther away from traffic sensors). If, at step 1215, it is determined that there are sufficient healthy traffic sensors, the routine continues to step 1220, where from other healthy traffic sensors in the road segment of the selected time category. Based on the data reading, determine a corrected data reading for an unhealthy traffic sensor. Corrected data readings can be determined in various ways, such as by calculating the average of multiple data readings obtained from healthy traffic sensors in the indicated road segment for the selected time category. it can. In some embodiments, all healthy traffic sensors can be used for averaging, while in other embodiments, only selected healthy traffic sensors can be used. For example, if a predetermined percentage (eg, at least 30%) of the traffic sensors in the indicated road segment is healthy during the selected time category, all healthy traffic sensors will be averaged. Otherwise, only a predetermined number (eg, at least two) of the closest healthy traffic sensors can be used.

  Instead, if it is determined in step 1215 that there are not enough healthy traffic sensors in the indicated road segment for the selected time category, the routine continues to step 1225, where: Attempts to determine an unhealthy road sensor corrected data read based on traffic sensors and / or other information related to the road segment. For example, such information may include predicted traffic condition information for road segments and / or unhealthy traffic sensors, predicted traffic condition information for road segments and / or unhealthy traffic sensors, and / or road segments and / or Or, historical average traffic condition information of unhealthy traffic sensors can be included. Different logic can be implemented to reflect the relative reliability of different types of information. For example, in some embodiments, predicted traffic situation information can be used in preference to predicted traffic situation information (eg, as long as it is available), and the predicted traffic situation information can be used as historical average traffic situation information. It can be used with higher priority. Additional details regarding prediction and forecasting of future traffic flow conditions are filed on March 3, 2006 and are entitled US Patent Application Serial No. 11 / 367,463 entitled “Dynamic Time Series Prediction Of Future Traffic Conditions”. Available and is incorporated herein by reference in its entirety. In other embodiments, the data read correction at step 1220 is necessarily performed based on the best data available from other healthy traffic sensors during the selected time category and / or associated time category. For example, steps 1215 and 1225 may not be performed. For example, data read correction may include all healthy traffic sensors in the indicated road segment of the selected time category when at least a predetermined percentage (eg, at least 30%) of all traffic sensors is healthy. To the nearest neighboring healthy traffic sensor in the indicated road segment and / or nearby road segment in the selected time category and / or related time category otherwise Can be based on.

  After step 1220 or 1225, the routine proceeds to step 1230 and provides the determined traffic sensor data reading for use as a corrected reading of traffic sensors in the selected time category. In some embodiments, the determined traffic sensor data reading is stored (eg, in a database or file system) for later use by other components (eg, sensor data aggregator component 110 of FIG. 1). Can do. At step 1235, the routine determines whether there are additional combinations of traffic sensors and time categories to process. If so, the routine returns to step 1210; otherwise, the routine proceeds to step 1299 and returns.

  FIG. 13 is a flowchart of an example embodiment of a sensor data read aggregator routine 1300. This routine may be used, for example, to determine and supply traffic status information for multiple traffic sensors during a particular time category or other time period, such as with respect to multiple traffic sensors associated with a particular road segment. Implementation of the sensor data aggregator component 355 and / or the sensor data aggregator component 110 of FIG. In the illustrated example embodiment, this routine is performed for a particular road segment, but in other embodiments, information from other types of groups of multiple traffic sensors can be aggregated. In addition, this routine may include other routines that perform an assessment of traffic situation information, such as to provide traffic situation information in situations where other routines cannot provide an accurate assessment (eg, due to insufficient data). Traffic status information complementary to the information provided by (eg, a data sample flow assessor routine) can be provided.

  The routine begins at step 1305 and receives an indication of one or more road segments and one or more time categories or other time periods. At step 1310, the routine selects the next road segment of the one or more indicated road segments, starting with the first indicated road segment. At step 1315, the routine obtains some or all of the available traffic sensor data readings taken during the time period (s) indicated by all traffic sensors associated with the road segment. Such information can be obtained, for example, from the sensor data conditioner component 105 of FIG. 1 and / or the sensor data conditioner component 353 of FIG. Specifically, this routine may in some cases be determined by the traffic sensor data reading of a traffic sensor determined to be healthy and / or supplied by the sensor data reading error collector routine of FIG. The corrected traffic sensor data reading of the traffic sensor determined to be unhealthy can be obtained.

  Next, at step 1320, the routine performs various data readings such as to determine the average speed, volume, and / or occupancy of the road segment during the indicated time period (s). Aggregate in one or more of the various forms. The average speed can be determined, for example, by taking an average of data readings reflecting the speed of the vehicle passing over one or more traffic sensors. Traffic volume can be determined with reference to data readings reporting vehicle counts. For example, given a loop sensor that reports the cumulative number of vehicles that have passed over the sensor since it was activated, traffic was simply obtained during the time period indicated. Inference can be made by subtracting two data reads and dividing the result by the time interval between the data reads. Further, the density can be determined based on the determined average speed, amount, and average vehicle length, as will be described in more detail elsewhere. In some cases, data reads can be weighted in various ways (eg, by age) so that more recent data reads have a greater impact on average flow determination than older data reads.

  Next, at step 1325, the routine determines whether there are more road segments (or other groups of traffic sensors) to process. If so, the routine returns to step 1310, otherwise it proceeds to step 1330 to provide the determined traffic flow information. In some embodiments, the determined flow information is stored for later supply to the traffic data client 109 of FIG. 1 and / or the RT information provider system 363 of FIG. 3 (eg, in a database or file system). Can) The routine then continues to step 1339 and returns.

  FIG. 14 is a flow diagram of an example embodiment of a traffic flow estimator routine 1400. This routine may be implemented by execution of a traffic flow estimator component (not shown), such as to estimate various types of traffic flow information for road segments in various ways. In this example embodiment, this routine is sufficient for the data sample rate assessor routine of FIG. 6 and / or the data sample flow assessor routine of FIG. 7 to provide sufficient data for these routines to accurately perform their respective evaluations. Can be called to obtain an estimate of quantity and / or occupancy, such as in situations where it is not otherwise available.

  The routine begins at step 1405 and displays a display of road segments, one or more time categories or other time periods, and one or more types of traffic flow information such as speed, quantity, density, occupancy. receive. At step 1410, the routine is based on whether one or more associated road segments have accurate information about one or more types of traffic flow information during one or more indicated time periods. Etc., to determine whether to estimate the indicated type of traffic flow information based on such road segments. Related road segments can be identified in various ways. For example, in some cases, the information about a road segment uses the traffic flow information of the second road segment, where the first road segment usually has a traffic pattern similar to the second (eg, adjacent) road segment. Information regarding the relationship between the road segments, such as being able to estimate traffic flow on the first road segment. In some cases, such a relationship is a statistical analysis of the respective traffic flow patterns on two road segments, whether previously analyzed and / or dynamically performed (eg, , Similar in shape to that previously described with respect to identifying similar data distributions for a given traffic sensor at different times, but instead, the similarity between multiple different traffic sensors such as the same time For example, based on analysis). Alternatively, one or more adjacent road segments may be selected as relevant for the indicated road segment without performing any particular relationship determination between the road segments. If it is determined to estimate the traffic flow information based on the associated road segment, the routine proceeds to step 1415 and the same type (s) of traffic flow of the associated road segment or segments. Based on the information, estimate the value (s) of traffic flow information of the indicated type (s). For example, the average speed of a road segment is based on the average traffic speed of one or more adjacent road segments (eg, by using the traffic speed from one adjacent road segment, or multiple adjacent roads Can be determined by taking the average of the traffic speed from the segment).

  If, in step 1410, it is instead determined not to estimate traffic flow information for the indicated road segment based on the associated road segment, the routine continues to step 1420 and the indicated road segment. And determining whether to estimate traffic flow information for the indicated road segment during the indicated time period or periods based on the predicted information for the indicated time period. In some embodiments, such predicted information is repeated for multiple future times (eg, 15 for the next 3 hours) while accurate current data is available. May be available only under certain conditions, such as every minute). Thus, if accurate input data to generate a forecast becomes available for an extended period of time (eg, over 3 hours), obtain a future traffic situation information forecast that can be utilized by this routine It may not be possible. Instead, in some embodiments, such predicted future traffic status information may not be available for other reasons, such as due to not being used in that embodiment. If it is determined at step 1420 to estimate traffic flow information based on the predicted information, the routine proceeds to step 1425 where the predicted information obtained, for example, from the predicted traffic information provider system 360 of FIG. Based on the indicated road segment and the indicated type (s) of traffic flow information for the indicated time period (s). Additional details regarding forecasting and forecasting of future traffic flow conditions are available in US Pat. No. 6,057,086, filed Mar. 3, 2006, entitled “Dynamic Time Series Prediction Of Future Traffic Conditions” and by reference. The entirety of which is incorporated herein.

  If step 1420 determines that the traffic flow information for the indicated segment is not estimated based on the predicted information (eg, due to the predicted information not being available), the routine Continuing to step 1430, the traffic flow information during the indicated road segment and one or more indicated time periods is estimated based on the forecast information for that road segment during that time period (s). Determine whether or not. In some embodiments, traffic conditions can be predicted for future times beyond the time at which traffic conditions are predicted, such as without using at least some current condition information. Thus, if the predicted information is not available (for example, because the exact input data to generate the prediction was not available for more than 3 hours), the forecast would still be significant, such as information that was generated in advance It may be possible to use information. If it is determined at step 1430 to estimate the traffic flow information based on the forecast information, the routine proceeds to step 1435 and is indicated, for example, based on the forecast information obtained from the forecast traffic information provider system 360. Estimate the indicated type (s) of traffic flow information for a given road segment and time period (s).

  Instead, if it is determined in step 1430 that the traffic flow information for the road segment indicated based on the forecast information is not estimated (eg, because forecast information is not available), this The routine continues to step 1440 and based on the historical average flow information for the indicated road segment (eg, for the same or corresponding time period, such as based on a time category including a time category and / or day of week category) Estimate the indicated type (s) of traffic flow information for the indicated road segment and time period (s). For example, if forecast information is not available (for example, because the input data is not available for a period longer than the period in which the most recent forecast and forecast were generated and cannot generate new forecasts and new forecasts) This routine can use historical average flow information for the indicated road segment. Additional details relating to the generation of historical average flow information are filed concurrently and are available in a US patent application entitled “Generating Representative Load Traffic Information From Historic Data” (Attorney Docket No. 480234.410P1). , Incorporated herein by reference in its entirety.

  After step 1415, 1425, 1435, or 1440, the routine proceeds to step 1445 for the indicated type (s) of the indicated road segment and the indicated time period (s). Provide estimated traffic flow information. The supplied information can be returned, for example, to the routine that called this routine (eg, a data sample flow assessor routine) and / or stored for later use (eg, in a database or file system). . After step 1445, the routine continues to step 1499 and returns.

  FIGS. 9A to 9C show examples of actions of a mobile data source when obtaining and providing information on road traffic conditions. Information regarding road traffic conditions is transmitted using radio links (eg, satellite uplink, cellular network, Wi-Fi, packet radio, etc.) and / or when the device reaches an appropriate docking point or other connection point In various forms, such as by physical download (for example, to download information from the fleet vehicle after it has returned to the main location or other destination with the appropriate equipment to perform the information download) It can be obtained from a mobile device (either a vehicle-based device and / or a user device). Information about road traffic conditions at the first time, obtained at a much later second time, has various benefits, such as what may be the case for information physically downloaded from the device. Provide (e.g., verification of predictions for the first time, use as observed case data in improving the prediction process later), but such road traffic situation information may be in real-time or near real-time form Provide additional benefits when obtained. Accordingly, in at least some embodiments, mobile devices with wireless communication capabilities may periodically (eg, every 30 seconds, every 1 minute, every 5 minutes, etc.) and / or with a sufficient amount of acquired information. When available (e.g., every acquisition of data points related to road traffic situation information, where N is a configurable number, etc., every N acquisitions of such data, storage with acquired data) At least some acquired information about road traffic conditions can be provided on a frequent basis, such as when the size and / or transmission size is reached. In some embodiments, such frequent wireless communication of acquired road traffic situation information includes additional data corresponding to each data point, such as to include aggregated information about multiple data points ( For example, during subsequent physical downloads from the device (via less frequent wireless communication with large amounts of data), it can be further supplemented by additional acquired road traffic situation information at other times. .

  Various benefits are provided by obtaining acquired road traffic situation information from mobile devices in real time or in other frequent ways, but in some embodiments, such road traffic situation information such as Wireless communication may be restricted in various ways. For example, in some cases, the cost structure of transmitting data from a mobile device over a particular radio link (eg, satellite uplink) occurs at less frequent intervals (eg, every 15 minutes). The mobile device may be pre-programmed to transmit at such intervals. In other cases, the mobile device is performed by the mobile device or the user of the device due to a lack of radio coverage in the area of the mobile device (eg, due to lack of a nearby cell phone receiver station) Due to other activity, the ability to transmit data over the wireless link may be temporarily lost, such as due to a temporary problem with the mobile device or associated transmitter.

  Thus, in some embodiments, at least some such mobile devices are designed to store multiple data samples (or have such multiple data samples stored on another associated device). Or configured in other ways, so that at least some information of multiple data samples can be transmitted together during a single wireless transmission. For example, in some embodiments, at least some mobile devices are in a period during which the mobile device cannot transmit data over the wireless link (eg, 30 seconds for mobile devices that typically transmit each data sample individually). Store the acquired road traffic situation information data samples at intervals (such as every other minute), and then generate these stored data samples together (or generate a subset and / or aggregation of these samples) It can be configured to transmit during wireless transmission. Some mobile devices can also be configured to perform wireless transmissions periodically (eg, every 15 minutes or when a specified amount of data is available for transmission), at least some implementations In a form, multiple data samples of road traffic situation information are acquired and stored (eg, at a predetermined sampling rate such as 30 seconds or 1 minute) over a time interval between wireless transmissions, and then Data samples may be further configured to be transmitted together (or a subset and / or aggregation of these samples) as well during wireless transmission. As an example, if wireless transmission of up to 1000 units of information costs $ 0.25 and each data sample is 50 units in size, it will be sampled every minute and 20 every 20 minutes It may be advantageous to send a data set containing a number of samples (rather than sending each sample individually every minute). In such embodiments, the data samples may be slightly delayed (in the periodic transmission example, assuming an average acquisition of data samples, on average, only half the time period between transmissions). However, the road traffic situation information obtained from this transmission still provides near real-time information. Further, in some embodiments, additional information can be generated and provided by the mobile device based on a plurality of stored data samples. For example, if a particular mobile device can only acquire information about the current instantaneous position in each data sample, but cannot acquire additional relevant information such as speed and / or position Related information can be calculated based on a plurality of subsequent data samples or otherwise determined.

Specifically, FIG. 9A shows an example area 955 having a plurality of interconnected roads 925, 930, 935, and 940, and a legend display 950 shows the north direction of the road (road 925 and 935 runs north-south and roads 930 and 940 run east-west). Although only a limited number of roads are shown, these roads may represent large geographic areas, such as interconnected freeways that span miles or a subset of streets that span many blocks . In this example, a mobile data source (eg, a vehicle not shown) has moved from position 945a to 945c over a period of 30 minutes and acquires and transmits a data sample indicating current traffic conditions every 15 minutes. Configured to do. Thus, when the mobile data source begins to move, it acquires and transmits the first data sample at location 945a (indicated by an asterisk “*” in this example), and after 15 minutes, at location 945b Two data samples are acquired and transmitted, and a third data sample is acquired and transmitted at position 945c after a total of 30 minutes. In this example, each data sample is represented as current position (eg, in GPS coordinates), current as represented using data values P _a , D _a , S _a , and T _a for 945a transmission. Includes indication of direction (eg, northbound), current speed (eg, 48 km / hour (30 miles)), and current time, and optionally other information (eg, identifier to indicate mobile data source) be able to. Such acquired and supplied current traffic situation information provides several benefits, but the route from location 945b to 945c occurred either partially along road 930 or along 940 Many details including these cannot be determined from such data. Further, such sample data does not allow, for example, the portion of road 925 between positions 945a and 945b to be treated as separate road segments that can report and predict separate traffic conditions.

  In a manner similar to FIG. 9A, FIG. 9B shows an example 905 having a mobile data source that travels from locations 945a to 945c on interconnected roads 925, 930, 935, and 940 over a period of 30 minutes, The mobile data source sends information about traffic conditions every 15 minutes (as indicated by the asterisks shown at locations 945a, 945b and 945c). However, in this example, the mobile data source is configured to acquire and transmit data samples every minute, and subsequent transmissions include data from each of the data samples in the previous 15 minutes. Thus, as the mobile data source moves between locations 945a and 945b, the mobile data source obtains a set 910b of 15 data samples 910b1 to 910b15, each data sample in this example being a data sample. Is shown with arrows pointing to the direction of the mobile data source. In this example, each data sample similarly includes an indication of the current position, current direction, current speed, and current time, and subsequent transmissions at position 945b include each of these data values of data sample 910b. Similarly, when the mobile data source moves between locations 945b and 945c, the mobile data source acquires 15 data samples 910c1-910c15, and subsequent transmissions at location 945c Contains each acquired data value of the data sample. By providing such additional data samples, various additional information can be obtained. For example, it can now be easily determined that the route from position 945b to 945c has partially occurred along road 930, not road 940, and the corresponding traffic status information can be returned to road 930. . Furthermore, the specific data samples and the data samples adjacent to them represent the road 925 between locations 945a and 945b, for example, as up to 15 separate road segments each having a distinct road traffic situation. Various information about smaller sections of the road can be provided, such as to enable (eg, by associating each data sample with a separate road segment). For example, the average speed of data samples 910b1-910b6 is approximately static (since the data samples are approximately equally spaced), and the average speed is increased for data samples 910b7 and 910b8 (these data samples are more distant Corresponds to a position, which in this example reflects a longer distance movement during a given 1 minute interval between data samples), and that the average velocity has been reduced for data samples 910b11-910b15 Can be observed. Although the data samples in this example provide information directly about such velocities, in other embodiments, such velocity information can be derived from data sample information that includes only the current position.

  FIG. 9C has a mobile data source that travels from location 965a to 965c on a portion of the interconnected road over a period of 30 minutes, where the mobile data source sends information about traffic conditions every 15 minutes (location 965a , 965b, and 965c (as shown by the asterisks shown), a third example 990 is shown. In FIG. 9C, the mobile data source is configured to acquire and store data samples every other minute in this example, and subsequent transmissions include data from each of at least some of the data samples in the previous 15 minutes. including. Thus, when the mobile data source moves between locations 965a and 965b, the mobile data source obtains a set 960b of 15 data samples 960b1-960b15. However, as indicated by co-located data samples 960b5-b13 (at this moment no movement was detected for these data samples so circles were used instead of arrows, but these circles In this example, the mobile data source has been stopped for about nine minutes at a location beside road 925 (eg, to stop at a coffee shop), shown separately from each other for clarity). . Thus, when the next transmission at location 965b occurs, the transmission may, in some embodiments, include all of the information regarding all of the data samples, or alternatively, at least such information. Some can be omitted (for example, in this situation, if the mobile data source is known to have remained immobile between data samples 960b5 and 960b13, data samples 960b6-960b12 are added Because it does not provide useful information, the information of data samples 960b6 to 960b12 is omitted). Further, although not shown in this figure, in other embodiments where the information of one or more such data samples is omitted, subsequent transmissions can use 15 data samples to be transmitted. (E.g., when periodic transmissions are performed based on the amount of data to be transmitted rather than time). In addition, when the mobile data source moves between locations 965b and 965c, the mobile data source is currently disabled for wireless communication (in this example shown with a white circle instead of an arrow). Data samples 960c13 and 960c14 are acquired in the area. In other embodiments, where each data sample is transmitted separately as it is acquired, but not otherwise stored, these data samples are lost, but in this example, instead of other data samples Stored with 960c1-960c12 and 930c15 and transmitted at location 965c. Although not shown in this figure, in some situations, the mobile data source may additionally temporarily lose the ability to obtain one or more data samples using the main means of data acquisition (E.g., if the mobile data source loses the ability to obtain GPS readings for 2-3 minutes), and if so, the mobile data source has been Data samples can be reported without further action (eg, to allow the receiver to interpolate or otherwise estimate these data samples if it is desired), but other In an embodiment, even if these data samples have lower accuracy or accuracy (e.g., this is better for these data samples). Can be reflected by including a low degree of confidence or a higher degree of possible errors, or by including other indications of how these and / or other data samples were generated) You can try to obtain data samples in other ways (for example, by determining the location, such as cell phone tower triangulation, by using a more inaccurate mechanism, or via dead reckoning) By estimating the current position based on the known position and subsequent average speed and direction of travel).

  Although the example data samples in each of FIGS. 9B and 9C are illustrated for a single vehicle or other mobile data source for clarity of illustration, in other embodiments a plurality of specific mobile data sources Data samples may not be used to determine the specific route that the mobile data source follows and, more generally, may not even be associated with each other (eg, each mobile data sample Source is anonymous or otherwise indistinguishable from other sources). For example, if multiple data samples from a particular mobile data source contain aggregated data associated with these data samples with each data sample or when such aggregated data is not used, When not used by a receiver to generate (eg, to generate velocity and / or position information based on consecutive data samples that provide only position information), such a receiver may have several implementations In form, identification data associated with a source of mobile data samples and / or multiple data samples may not be provided with an indication that they are from the same mobile data source (eg, mobile data Based on design decisions to increase source related privacy Stomach).

  Instead, in at least some such embodiments, multiple mobile data sources all mobile data for that road segment to determine aggregate information for a particular road segment (or other part of the road) Used together to determine interesting road condition information, such as to use multiple data samples from a source. Thus, for example, during a period of interest (eg, 1 minute, 5 minutes, 15 minutes, etc.), each of a number of unrelated mobile data sources may be on a particular road segment during that time period. One or more data samples related to its own movement of the sensor, and if each such data sample contains speed and direction information (for example), the average aggregate speed For all mobile data sources that are moving in the same direction as a whole, such as in a form similar to a road sensor that aggregates information for multiple vehicles passing through the vehicle, the time period and the road segment can be determined. A specified data criterion, such as a road (or road segment) with a location closest to the data sample location (for any road, or having one or more indicated functional road types) Supplied by a mobile data source by selecting the appropriate road segment for that road, or with a data sample of the relevant road (or road segment) It can be associated with a particular road segment in various ways, such as by using a display. Furthermore, in at least some embodiments, roads other than one-way roads are assigned to data samples to roads and for other purposes (eg, the freeway northbound lane is distinct from the freeway southbound lane). Treated as a separate road, and if so, the direction of the mobile data sample can be used to further determine the appropriate road associated with the data sample, while others In this embodiment, to treat a bi-directional street as a single road (eg, with average traffic conditions reported and predicted for vehicles traveling in both directions), each lane or other road in a multi-lane freeway is separately Roads can be modeled in other ways, such as treating them as logical roads.

  In some embodiments, fleet vehicles are provided to provide useful data samples to facilitate the use of multiple mobile data sources to determine interested road condition information. It can be configured in various forms. For example, if a large fleet of vehicles depart each day from the same origin point at a similar time every day, various fleet vehicles all minimize a very large number of data points near a single origin point Differently in terms of how quickly and how often to start supplying data samples, such as and / or to provide variability when data samples are acquired and transmitted. More generally, the mobile data source device is defined as the total distance covered since the starting point (eg, the starting point of the fleet vehicle group), the distance covered since the last data sample acquisition and / or transmission, the starting time. The indicated relationship that has occurred for one or more indicated positions, for example, the total time that has elapsed since (eg, departure of the fleet vehicle from the origin point), the time that has elapsed since the last data sample acquisition and / or transmission It can be configured in various ways regarding how and when to acquire data samples, including based on (eg, transit, arrival, departure, etc.). Similarly, the mobile data source device may have a total distance covered since the start point, distance covered since the last data sample acquisition and / or transmission, total time elapsed since the start time, last data sample acquisition and / Or the time elapsed since transmission, the indicated relationship that occurred for one or more indicated locations, the indicated number of collected data samples, the indicated amount of collected data (e.g., mobile The amount used to store data samples on the device, such as full or substantially full, or the indicated length of time for transmission to be full or substantially full One or more acquired data samples such as when a predefined condition is satisfied, including It can be configured in various ways with respect to either deliver at any time send or otherwise as.

  8 may be, for example, the vehicle-based data source 384 of FIG. 3 and / or the other data source 388 (eg, user device) of FIG. 3 and / or the vehicle-based data source 101 of FIG. 7 is a flow diagram of an example embodiment of a mobile data source information provision routine 800, such as can be achieved by operation of each mobile data source device of one or more of the other data sources 102. In this example, the routine obtains data samples from a particular mobile data source to indicate current traffic conditions and appropriately samples the data samples so that subsequent transmissions can include information from multiple data samples. Store.

  The routine begins at step 805, where configurable parameters indicate when data samples must be acquired and when transmission with information corresponding to one or more data samples must be performed, etc. Retrieve parameters used as part of acquiring and supplying data samples. The routine continues to step 810, where the retrieved parameters and / or other information (eg, the indicated length of time since the previous data sample acquisition, the indication moved since the previous data sample acquisition). Wait until it is time to acquire a data sample, such as an indication of acquiring the data sample in a substantially continuous manner). The routine then continues to step 815, obtaining a data sample based on the current location and movement of the mobile data source, and storing the data sample at step 820. At step 825, the retrieved parameters and / or other information (eg, that the indicated length of time has elapsed since the previous transmission, that the indicated distance has been moved since the previous transmission, data samples, If it is determined that it is not time to send data yet (eg, based on an indication of sending data samples in a substantially continuous manner) Return to step 810.

  Otherwise, the routine continues to step 830 to retrieve and select all stored data samples since the previous transmission (or since startup in the case of the first transmission). The routine then optionally generates aggregated data based on a plurality of selected data samples (eg, the overall average rate of the data samples, acquired information, step 835). (Eg, the average speed and direction of each data sample when providing only location information) may not perform such aggregated data generation in other embodiments. Next, at step 840, the routine optionally removes some or all of the acquired information about some or all of the data samples from the selected set of data samples (eg, for each data sample). To send only selected types of data, to remove data samples that appear to be outliers or otherwise in error, to remove data samples that do not correspond to the actual movement of the mobile data source In other embodiments, such information removal may not be performed. Next, at step 845, the routine sends the current information in the current set of data samples and all aggregated information to the receiver, which uses the data in an appropriate manner. At step 895, the routine determines whether to continue (eg, whether the mobile data source continues to be used and remains mobile), and if so, returns to step 810. Otherwise, the routine continues to step 899 and ends. In embodiments and situations where the mobile data source cannot send data, either due to temporary conditions or instead to reflect the configuration or restrictions of the mobile data source, the mobile data source has not been sent since the previous transmission. Steps 830-845 may not be executed until some or all of the acquired and stored data samples can be transmitted or otherwise provided (eg, via physical download). is there.

  As previously noted, once information about road traffic conditions is obtained, such as from one or more mobile data sources and / or one or more other sources, the road traffic condition information is substantially real-time. Used in various forms, such as to report current road traffic status in the form of or to use past and current road traffic status information to predict future traffic status at each of multiple future times can do. In some embodiments, the type of input data used to generate future traffic forecasts includes various current, past, and expected future conditions, as described in more detail elsewhere. And the output from the prediction process can include each of a plurality of future times within a predetermined time interval (eg, 3 hours or 1 day) (eg, every 5 minutes in the future, 15 Generated predictions of expected traffic conditions at each of a plurality of target road segments of interest (every minute or every 60 minutes) may be included. For example, the type of input data includes information about the current and past amount of traffic for various target road segments of interest within the geographic area, such as a network of selected roads within the geographic area, And information on recent traffic accidents, information on current, recent, and future road construction, information on current, past, and expected future weather conditions (eg, precipitation, temperature, wind direction, wind speed, etc.), at least Information about current, past, and future scheduled events (eg, all events, type of event shown, exceeded the indicated threshold (eg, 1000 or 5000 expected attendees)) The type of event, such as an event that is large enough, such as having expected attendees, The expected start and end time of the vent and / or event venue or other location) and information about the school schedule (eg, whether the school is in class and / or one or more of the schools) Position). Further, in some embodiments, the multiple future times at which future traffic conditions are predicted are at each point in time, but in other embodiments, such predictions are instead in multiple points in time. Such multiple points in time (eg, a period of time) can be represented, such as by representing an average value or other aggregated measurement of future traffic conditions. In addition, some or all of the input data can be known and expressed with varying accuracy (eg, expected weather), generating additional information to generate confidence in the generated prediction and / or other Metadata can be represented. In addition, predictions of future traffic conditions can be made periodically (eg every 5 minutes), in response to user requests when any or enough new input data is received, etc. You can start at various times for a reason.

  In some embodiments, a portion of the same type of input data can be used to generate a long-term forecast of future traffic conditions (eg, one week ahead or one month ahead) as well, Such long-term forecasts may not use some of the types of input data, such as information about the current state at the time of forecast generation (eg, current traffic conditions, weather conditions, or other conditions). is there. In addition, such long-term forecasts can be generated less frequently than short-term forecasts and are made to reflect future time periods that differ from short-term forecasts (eg, every hour instead of every 15 minutes). There is.

  The roads and / or road segments for which future traffic forecasts and / or traffic forecasts are generated can also be selected in various ways in various embodiments. In some embodiments, future traffic forecasts and / or traffic forecasts are generated for each of a plurality of geographic areas (eg, metropolitan areas), each geographic area having a plurality of interconnected roads. Such a geographic area is an area where current traffic status information is readily available (eg, based on a network of road sensors of at least some of the roads in the area) and / or traffic You can choose in a variety of ways, such as based on areas where congestion is a big problem. In some such embodiments, roads for which future traffic forecasts and / or traffic forecasts are generated include roads for which current traffic situation information is readily available, while in other embodiments The selection of such roads is based at least in part on one or more other factors (e.g., based on road size or capacity, such as to include freeways and major highways, etc. Road functions, such as those specified by the US Federal Road Administration, based on the role the road plays in carrying traffic, such as to include main roads and collector roads, which are major alternatives to higher capacity roads For example). In other embodiments, future traffic situation predictions and / or traffic situation forecasts may be made for a single road regardless of road size and / or correlation with other roads. Further, in order to treat future traffic conditions and / or road segments for which traffic condition predictions are generated as a separate segment for each road sensor, multiple road sensors are grouped together by road segment (eg, To reduce the number of independent predictions and / or forecasts made, such as by grouping together a specified number of road sensors), traffic conditions information from traffic sensors and / or other sources (e.g., Traffic conditions are usually the same or sufficiently similar (eg, strongly correlated), such as based on vehicles traveling on the road and / or data generated by the user, as described in more detail elsewhere. ) Select in various ways, such as to select road segments to reflect logically related sections of roads It can be.

  In addition, future traffic forecast information and / or traffic forecast information may vary at various times (eg, by periodically sending information in response to a request), as described in more detail elsewhere. Information analysis and / or by displaying the information to the user in a form (e.g., via a web browser and / or application program, such as by sending the information to a cell phone and / or other portable consumer device) Or providing such information to the user and / or organization (such as by providing information to other organizations and / or entities that provide at least a portion of the information to the user, such as a third party that provides the information after the change) Used in various forms in various embodiments, including for supply It can be. For example, in some embodiments, the forecast information and / or forecast information is used to optimally perform an optimal route and / or indicated movement between a start location and an end location through a network of roads. A proposed travel route and / or time, such as time, can be determined, such determination can include prediction information at each of a plurality of future times for one or more roads and / or road segments and / or Or based on forecast information.

  Further, various embodiments allow users and other clients to use one or more of the traffic information systems (eg, data sample manager system 350, RT information provider system 363, and / or predictive traffic information provider system of FIG. 3). 360) to provide various mechanisms to interact with. For example, some embodiments may allow a client to analyze, select, and / or provide information requests related to current and / or predicted traffic conditions and / or information related to travel routes. An interactive console (eg, a client program that provides an interactive user interface, a web browser-based interface, etc.) that can make requests and receive corresponding responses can be provided. Further, some embodiments allow client computing systems to programmatically perform some or all of such requests, such as via network message protocols (eg, web services) and / or other communication mechanisms. Provides an API ("Application Programmer Interface") that enables it.

  Those skilled in the art will provide, in some embodiments, alternatives such as dividing the functionality provided by the routines described above into more routines or consolidating them into fewer routines. You will understand what you can do. Similarly, in some embodiments, the indicated routine, such as when other indicated routines instead lack or include each such functionality or when the amount of functionality provided is changed. Can provide more or fewer functionalities than those described. Moreover, while various operations have been illustrated as being performed in a particular fashion (eg, in series or in parallel) and / or in a particular order, those skilled in the art will recognize other operations in other embodiments. It will be appreciated that it can be performed in order and in other ways. Those skilled in the art can structure the data structures described above in different ways, such as by dividing a single data structure into multiple data structures or by consolidating multiple data structures into a single data structure. You will understand that Similarly, in some embodiments, such as when other illustrated data structures instead lack or contain such information, respectively, or when the amount or type of information stored is changed, etc. A data structure can store more or less information than described.

  From the foregoing, it will be understood that although particular embodiments have been described herein for purposes of illustration, various modifications may be made without departing from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims and the elements recited therein. Furthermore, while certain aspects of the invention are described in certain claim forms, the inventors contemplate the various aspects of the invention in all available claim forms. For example, only some aspects of the invention may currently be listed as being implemented on a computer-readable medium, but other aspects can be practiced as such.

FIG. 3 is a block diagram illustrating data flow between components of an embodiment of a system that evaluates road traffic conditions based at least in part on data obtained from vehicles and other mobile data sources. FIG. 6 illustrates an example of road traffic situation assessment based at least in part on data obtained from vehicles and other mobile data sources. FIG. 6 illustrates an example of road traffic situation assessment based at least in part on data obtained from vehicles and other mobile data sources. FIG. 6 illustrates an example of road traffic situation assessment based at least in part on data obtained from vehicles and other mobile data sources. FIG. 6 illustrates an example of road traffic situation assessment based at least in part on data obtained from vehicles and other mobile data sources. FIG. 6 illustrates an example of road traffic situation assessment based at least in part on data obtained from vehicles and other mobile data sources. FIG. 2 is a block diagram illustrating a computing system suitable for executing embodiments of the described data sample manager system. 6 is a flow diagram illustrating an example embodiment of a data sample filter routine. 6 is a flow diagram illustrating an example embodiment of a data sample outlier eliminator routine. 5 is a flow diagram illustrating an example embodiment of a data sample rate assessor routine. 5 is a flow diagram illustrating an example embodiment of a data sample flow assessor routine. 5 is a flow diagram illustrating an example embodiment of a mobile data source information provision routine. It is a figure which shows the example of the action of a mobile data source at the time of acquiring and providing the information regarding a road traffic condition. It is a figure which shows the example of the action of a mobile data source at the time of acquiring and providing the information regarding a road traffic condition. It is a figure which shows the example of the action of a mobile data source at the time of acquiring and providing the information regarding a road traffic condition. It is a figure which shows the example of correction of the data sample acquired from the road traffic sensor. It is a figure which shows the example of correction of the data sample acquired from the road traffic sensor. 6 is a flow diagram illustrating an example embodiment of a sensor data read error detector routine. 6 is a flow diagram illustrating an example embodiment of a sensor data read error collector routine. 6 is a flow diagram illustrating an example embodiment of a sensor data read aggregator routine. 6 is a flow diagram illustrating an example embodiment of a traffic flow estimator routine.

Claims (310)

A computer-implemented method for determining estimated average speed information for vehicles traveling on roads based on data samples reflecting movement on these roads, comprising:
Receiving an indication of one or more segments of one or more roads, each road segment comprising a number of vehicles each reflecting a reported speed of a vehicle on the road segment at a reported time. A step with associated data samples;
Automatically estimating, for each of at least one of the road segments, an average traffic speed of a vehicle traveling on the road segment during a period of time,
Identifying the group of multiple data samples associated with the road segment whose reported time occurs during the period of time;
Determining the weight of the data samples based on one or more attributes of the data samples that affect the accuracy of the reported rate of the data samples of the group;
Determining the estimated average traffic speed of vehicles traveling on the road segment during the period of time based at least in part on a weighted average of the reported speeds of the data samples of the group. The weighted average is calculated using the determined weight and is automatically estimated by determining;
Using one or more of the estimated average traffic speeds to facilitate travel on the one or more roads. The received indication of the one or more segments of the one or more roads includes an indication of multiple road segments of the one or more roads, each of the at least one road segment during a period of time. The automatic estimation of the average traffic speed of the vehicle traveling at is performed for each of a number of time periods, the number of data relating to at least one road segment whose reported time occurs during the time period Identifying the group of samples is
Receiving information related to current traffic conditions of the multiple road segments during the period of time, wherein the received information is each reported from one of a number of vehicles, during the period of time. Including a plurality of data samples for the period of time reflecting the reported speed of the one vehicle at a reported location on one of the road segments at a reported time of Is a subset of all vehicles traveling on the road segment during the period of time, and the received information is each reported from one of a number of traffic sensors monitoring the number of road segments. One or more of one or more vehicles at a position above one of the road segments at one or more reported times during the period of time. Reported to reflect the rate is based on the speed reading, further comprising a plurality of additional data samples of said predetermined period, to receive and,
Identifying, for each of the multiple road segments, a group of multiple data samples of the road segment for the period of time, the multiple data samples comprising the plurality of data samples and the plurality of additional data. Identifying the one or more attributes of the data samples of the group to which the weight of the data samples is determined, wherein the one or more attributes are from which at least one of the samples is identified. Data samples with less recent times are given less weight than data samples whose reported times are more recent, and data samples reported from vehicles on the one or more roads are Data samples so that they are given different weights than reported data samples Including the freshness of the reported time and source of data samples,
As a result, average traffic speed is calculated for road segments based on data samples that reflect actual vehicle movement on the road segments that are weighted to reflect the freshness of the data samples and the source of the data samples. The method of claim 1, wherein the method is determined.   For each of at least one of the time periods and at least one of the road segments, the identified group of a number of data samples of the road segments of the time period has a reported position of the Including a number of data samples from the plurality of data samples of the time period on the road segment, the position of the plurality of data samples from the plurality of additional data samples of the time period being on the road segment. The method of claim 2 comprising:   For each of at least one of the time periods and at least one of the road segments, the determination of the weight of each of the data samples of the group further expects its reported speed to be lower The expected accuracy of the reported rate of the data sample so that the data sample with the reported accuracy is given a weight less than the data sample with the higher expected accuracy of the reported rate The method of claim 2, wherein the method is based at least in part.   For the at least one fixed period and the at least one road segment, the determination of the weight of each of the data samples of the group is determined by determining whether the data sample having the expected accuracy is lower in its reported speed. 5. The method of claim 4, wherein the reported rate is performed using an exponential weighting function such that the reported rate is given an exponentially smaller weight than a data sample having a higher expected accuracy. the method of.   For each of at least one of the time periods and each of at least one of the road segments, at least some of the data samples are used for multiple time windows and are determined differently for these time windows. For each of the time windows using the data samples whose reported time is in the time window, and dividing the fixed period into a number of overlapping time windows The method of claim 2, further comprising: performing the estimation of the average traffic speed of a vehicle traveling on the road segment during the period of time.   For each of at least one of the time periods and at least one of the road segments, the determination of the estimated average traffic speed of a vehicle traveling on the road segment during the time period is Generating the estimated average traffic speed confidence value to reflect the degree of possible error in the estimated average traffic speed, the one or more estimated average traffic for the period of time. The method of claim 2, wherein the use of speed comprises using the generated confidence value to facilitate future travel on the one or more roads.   The estimation of the average traffic speed of a vehicle traveling on a road segment during a period of time is performed using recently received data samples in real time, and the one or more estimates of the period of time are The use of average traffic speed is further performed in a substantially real-time manner to facilitate imminent future travel on the one or more roads. The method described.   The use of the one or more estimated average traffic speeds for a period of time to facilitate travel on the one or more roads is based in part on the estimated average traffic speeds. Inferring the traffic volume of at least one road segment of the one or more roads and the estimated average for one or more persons considering the upcoming movement on the one or more roads 9. The method of claim 8, comprising providing information regarding traffic speed and the inferred traffic volume.   For each one or more of the at least some road segments, the one or more attributes of the data samples of the group used in the determination of the weights are the reported of the data samples. The method according to claim 1, further comprising: newness of time.   The determination of the weight of a data sample based on the newness of the reported time of the data sample is less than a data sample whose data time is less recent than a data sample whose time reported is more recent The method according to claim 10, wherein the method is performed to be weighted.   For each one or more of the at least some road segments, the data samples associated with the road segment include data samples provided from a number of sources and are used in the determination of the weights. The method of claim 1, wherein the one or more attributes of the data samples of the group include the source of the data samples.   For each of at least one of the one or more road segments, one of the multiple sources of the data sample is moving over the road segment, and the data sample is based on the movement. One or more vehicles to report, and one of the multiple sources of the data sample monitors the road segment and reports the data sample based on a reading reflecting the passing vehicle 13. The method of claim 12, comprising one or more traffic sensors.   For each of at least one of the one or more road segments, the multiple sources of the data samples are moving over the road segment, each to reflect a vehicle location and / or speed. The method of claim 12, comprising: the plurality of vehicles reporting one or more data samples.   For each of at least one of the one or more road segments, the multiple sources of the data samples are based on readings that monitor the road segments, each reflecting one or more passing vehicles. The method of claim 12, including a number of traffic sensors, each reporting one or more data samples.   Further comprising, for each of at least one of the one or more road segments, evaluating the reliability of each of the plurality of sources of the data samples of the road segment; The determination of the weight of the data sample based is such that a data sample whose source has a lower rated reliability is given a smaller weight than a data sample whose source has a higher rated reliability. The method of claim 12, wherein the method is performed.   For each of one or more of the at least some road segments, the one or more attributes of the data sample of the group used for the determination of the weight are: The method of claim 1, comprising a total amount.   For each of one or more of the at least some road segments, the determination of the weight of the data samples of the group determines that the data sample reflects its expected accuracy of lower attributes. The method of claim 1, wherein the method is performed such that the attribute is weighted less than a data sample that reflects a higher expected accuracy.   The method of claim 1, wherein, for each of one or more of the at least some road segments, the determination of the weight of the data samples of the group uses exponential weighting. .   The determination of the weight of the data samples of the group for each of one or more of the at least some road segments is further based on one or more current conditions. Item 2. The method according to Item 1.   The current state includes at least one of a recent traffic accident, a sports event, a current time, a current day of the week, a current date, a current week, and a current month. The method described.   For each one or more of the at least some road segments, the data samples associated with the road segments of the identified group are moving over the road segments during the period of time; A data sample supplied from one or more vehicles that reports the data sample based on the movement, wherein the one or more vehicles are all moving on the road segment during the period of time. The estimated average traffic speed of a vehicle that is a subset of vehicles and travels on the road segment during the period is the average traffic speed of all the vehicles that travel on the road segment during the period The method of claim 1, wherein the method is determined in an estimated manner.   For each of one or more of the at least some road segments, the estimate of the average traffic speed of vehicles traveling on the road segment during the fixed period is for each of a number of separate fixed periods. The method of claim 1, wherein the method is performed.   For each of one or more of the at least some road segments, the estimate of the average traffic speed of a vehicle traveling over the road segment over a period of time is obtained from the associated data sample of the road segment. 2. The method of claim 1, wherein the method is performed for each of a number of overlapping time windows during the time period such that at least some of them are used for each of a number of the time windows. the method of.   For at least one of the one or more road segments, the multiple overlapping time windows during the period of time are modified to reflect one or more current conditions. 25. The method of claim 24.   For each of one or more of the at least some road segments, the estimate of the average traffic speed of a vehicle traveling on the road segment during the period is a confidence in the estimated average traffic speed. Determining the degree value, wherein the use of the one or more estimated average traffic speeds determines the determined reliability to facilitate future travel on the one or more roads. The method of claim 1, comprising using one or more of the values.   The use of the one or more estimated average traffic speeds infers traffic volume of at least one of the road segments based in part on the one or more estimated average traffic speeds. The method of claim 1, comprising:   The use of the one or more estimated average traffic speeds may facilitate the one or more estimates to facilitate judgment by one or more persons regarding travel on the one or more roads. The method of claim 1 including providing the person with an indication of the average traffic speed being played.   The determination of the one or more estimated average traffic speeds and the use of the one or more estimated average traffic speeds to enable a substantially real-time determination by the person; 29. The method of claim 28, wherein the method is performed in a substantially substantially real-time manner relative to receiving the data samples used for determination.   The method of claim 1, wherein the automatic estimation of at least one of the estimated average traffic speeds is performed in a substantially real-time manner.   For each one or more of the at least some road segments, moving over the road segment during the period of time based at least in part on the weighted average of the reported speeds of the data samples of the group The determination of the estimated average traffic speed of a vehicle that determines whether the identified group of data samples associated with the road segment has a statistical validity of the weighted average of the reported speed threshold. 2. The method of claim 1, wherein the method is performed only if it is large enough to exceed. A computer readable medium, the contents enabling a computing device to estimate average speed information of a moving vehicle by performing the method, the method comprising:
Receiving an indication of a number of data samples each reflecting a speed of a number of vehicles traveling on the road;
Estimating an average traffic speed of a vehicle traveling on the road based at least in part on combining the speeds of the data samples in a weighted manner using weights associated with the data samples; Estimating the associated weights including a number of distinct weights;
Providing an indication of the estimated average traffic speed used to facilitate travel on the road.   The multiple data samples each have a reported time associated with the speed, and the estimated average traffic speed relates to a vehicle traveling on the road during a period of time including the reported time. The estimation of the average traffic speed of a vehicle traveling on the road determines the weight associated with each of the data samples in a manner that reflects the expected accuracy of the speed of the data samples. The combination of the speeds of the data samples in a weighted form includes determining a weighted average of the speeds of the data samples based on the determined weight. 33. The computer-readable medium according to 32.   The multiple data samples are reported by the multiple vehicles traveling on the road, wherein the multiple vehicles are a subset of all vehicles traveling on the road, and the determined weight is the multiple Represents the average traffic speed of the vehicle, and the estimated average traffic speed represents the average traffic speed of all the vehicles traveling on the road and is used in facilitating movement on the road The providing of the indication of estimated average traffic speed includes presenting the estimated average traffic speed to a vehicle driver for use in influencing decisions regarding travel on the road. The computer-readable medium of claim 33.   The computer-readable medium of claim 32, wherein the computer-readable medium is a memory of a computing device.   The computer-readable medium of claim 32, wherein the computer-readable medium is a data transmission medium that transmits a generated data signal including the content.   The computer-readable medium of claim 32, wherein the content is instructions that when executed cause the computing device to perform the method. A computing system configured to estimate average speed information of a moving vehicle, comprising:
A first component configured to receive, for each of a number of roads, an indication of a number of data samples associated with the road, each reflecting a speed of a vehicle traveling on the road;
For each of the multiple roads,
Determining the weight of these data samples based on one or more attributes of the data samples associated with the road;
A weighted average of the velocities of the data samples associated with the road on the road at one or more times based at least in part on the weighted average calculated using the determined weight Estimate the traffic speed of vehicles moving
A data sample speed assessor component configured to provide an indication of the estimated traffic speed for use in facilitating travel on the road.   For each of at least one of the multiple roads, each of the multiple data samples associated with the road has a reported time associated with the speed, and the estimated traffic speed is the report For a vehicle traveling on the road during a certain period of time including the determined time, the weight of the data samples associated with the road is determined in a manner that reflects the expected accuracy of the speed of these data samples And providing the indication of the estimated traffic speed for use in facilitating travel on the road is a vehicle driver for use in influencing decisions regarding travel on the road 40. The computing system of claim 38, comprising presenting the estimated traffic speed to a computer.   40. The computing system of claim 38, wherein the first component and the data sample rate assessor component each include software instructions that are executed in a memory of the computing system.   The first component comprises means for receiving, for each of a number of roads, a display of a number of data samples associated with the road, each reflecting a speed of a vehicle traveling on the road, the data sample speed assessor component Is calculated for each of the multiple roads using the determined weights, determining weights of these data samples based on one or more attributes of the data samples associated with the roads Estimating a traffic speed of a vehicle traveling on the road at one or more times based at least in part on a weighted average of the speeds of the data samples associated with the road to facilitate movement on the road Means for providing an indication of the estimated traffic speed for use in The computing system of claim 38 that. A computer-implemented method for automatically providing data samples from vehicles traveling on these roads for use in assessing traffic conditions on the roads, comprising:
For each of a number of vehicles traveling on one or more roads, for each of a number of consecutive periods of a predetermined length,
Acquiring a number of data samples during the time period from a geolocation device moving with the vehicle, each data sample being acquired at a separate acquisition time during the time period; Acquiring, including the reported geographical location and speed of the vehicle;
Temporarily storing the multiple data samples during the period of time in a storage device;
At the end of the period,
Via a wireless data transmitter to a remote traffic information system configured to use at least some of the data samples to facilitate movement of other vehicles on the one or more roads. Providing information on movement of the vehicle by transmitting together the multiple stored data samples in one data transmission;
Removing the multiple stored data samples from the storage device to allow temporary storage of data samples acquired during the next period of time.   43. The method of claim 42, wherein for each of the multiple vehicles, the provision of the information regarding the movement of the vehicle is performed by one or more computing systems on the vehicle. All of the multiple vehicles transmit data samples to a single remote traffic information system, and the method is further under the control of the single remote traffic information system,
Repeatedly receiving data transmissions from the multiple vehicles, each of the received data transmissions being acquired at a separate acquisition time, the geographical location and speed of the vehicle at those acquisition times Repeatedly receiving, including a large number of data samples reporting
For each of a number of road segments and a number of time periods of the one or more roads,
Retrieving the data samples received from the multiple vehicles whose acquisition time is during the time period and whose reported geographical location corresponds to the road segment;
Evaluating an average road travel condition of all vehicles traveling on the road segment during the period of time based at least in part on the reported speed of the retrieved data sample;
Using the estimated average road movement condition to assist movement of other vehicles on or near the road segment during or near the time period. 44. The method of claim 43.   The use of the estimated average road movement condition to assist in the movement of another vehicle sends information regarding the estimated average road movement condition to the other vehicle in substantially real-time manner; 45. Enabling at least one of information regarding future road movement conditions predicted based at least in part on the estimated average road movement conditions. the method of.   For each of at least some of the multiple vehicles, the geolocation device that moves with the vehicle is a Global Positioning System (“GPS”) receiver installed within the vehicle. 44. The method of claim 43.   44. The method of claim 43, wherein for each of at least some of the multiple vehicles, the geolocation device that moves with the vehicle is a cell phone carried by passengers in the vehicle. .   For each of at least some of the multiple vehicles, the acquired data samples of the vehicles are respectively reported travel directions of the vehicle at the acquisition time of the data samples, and the data samples of the data samples. Display of device status of the geolocation device moving with the vehicle at acquisition time, display of vehicle status of the vehicle at the acquisition time of the data sample, and associated with the geolocation device and / or the vehicle 44. The method of claim 43, further comprising a unique identifier.   For each of the at least some multiple vehicles, the wireless data transmitter used for the transmission of multiple data samples is a satellite transmitter that transmits only a limited amount of data, and the at least some For each of a number of vehicles, the transmission of a number of data samples together in a single data transmission via the wireless transmitter may reduce the amount of data contained in the single data transmission. 49. The method of claim 48, comprising processing the multiple data samples prior to a data transmission.   For one of the at least some number of vehicles, the providing of the information regarding the movement of the one vehicle during one of the time periods may include one or more additional data samples. The one or more additional data samples of the one vehicle during the one period so as to be distinct from the multiple data samples acquired during the one period of the one vehicle The processing of the multiple data samples prior to the single data transmission of the one vehicle and the one period of time includes moving the additional data samples of interest. 50. determining to exclude the additional data sample from the single data transmission based on determining that it does not reflect The method described.   For each of at least some of the multiple vehicles, the predetermined length of each of the fixed periods is 15 minutes and the data samples are acquired at intervals of about 1 minute. 43. The method of claim 42. A method for a device in a vehicle that automatically provides information regarding movement of the vehicle on one or more roads, under the control of the device,
Obtaining a number of data samples each acquired at the acquisition time to report the geographical location and movement characteristics of the vehicle at separate acquisition times;
Storing the multiple data samples until one or more predefined conditions are satisfied;
A single traffic information system configured to use data samples provided from multiple vehicles to facilitate travel on the road after the one or more predefined conditions are satisfied Automatically providing said multiple data samples in transmission.   53. The method of claim 52, wherein the one or more predefined conditions include a indicated length of a period of time during which the obtaining of the multiple data samples is performed.   54. The method of claim 53, wherein the indicated length of time is at least 10 minutes and the acquisition time of the data samples occurs about every minute.   The obtaining of a number of data samples, the storage of the number of data samples, and the provision of the number of data samples in a single data transmission, respectively, for a number of time periods of the indicated length 54. The method of claim 53, wherein the method is repeatedly performed to provide information regarding movement of the vehicle over an extended period of time including:   53. The method of claim 52, wherein the one or more predefined conditions include a indicated time at which the provision of the multiple data samples in a single data transmission must be performed. Method.   53. The method of claim 52, wherein the one or more predefined conditions include a indicated number of multiple data samples that must be obtained.   53. The method of claim 52, wherein the one or more predefined conditions include a indicated amount of data that must be obtained from the multiple data samples.   53. The method of claim 52, wherein the one or more predefined conditions include a indicated distance that the vehicle has traveled.   53. The method of claim 52, wherein at least some of the multiple data samples are each acquired after the indicated length of time has elapsed since the previous data sample was acquired. the method of.   53. The method of claim 52, wherein at least some of the multiple data samples are each acquired after the vehicle has traveled the indicated distance since the previous data sample was acquired. the method of.   53. The method of claim 52, wherein at least some of the multiple data samples are each acquired after the vehicle has arrived at one or more indicated geographic locations.   The acquisition of multiple data samples, the storage of the multiple data samples, and the provision of the multiple data samples in a single data transmission are each performed repeatedly. Item 53. The method according to Item 52.   53. The method of claim 52, wherein at least some of the data samples report the geographic location of the vehicle using latitude and longitude coordinates, respectively.   53. The method of claim 52, wherein at least some of the data samples each report the geographic location of the vehicle using a location based on a road on which the vehicle is moving. .   53. The method of claim 52, wherein the reported movement characteristics of each vehicle of at least some of the data samples includes a speed of the vehicle.   53. The method of claim 52, wherein the reported movement characteristics of each vehicle of at least some of the data samples includes a direction of travel of the vehicle.   The reported travel characteristics of each of the vehicles of at least some of the data samples includes at least one of a distance traveled by the vehicle and a status of the vehicle. 52. The method according to 52.   At least some of the data samples are each associated with an indication of the status of the geolocation device used to obtain the geographic location of the data sample and the geolocation device and / or the vehicle 53. The method of claim 52, further comprising at least one of indications of identifiers to perform.   53. The method of claim 52, wherein the device in the vehicle includes a global positioning system ("GPS") receiver that provides geographical location information.   53. The method of claim 52, wherein the device in the vehicle includes a cell phone carried by passengers in the vehicle.   72. The method of claim 71, wherein the geographic location of each of at least some of the data samples is based on one or more cell phone network transmitters with which the cell phone can communicate.   72. The provisioning of the multiple data samples in the single data transmission includes transmitting the single data transmission over a cell phone network of the cell phone. the method of.   53. The method of claim 52, wherein the device in the vehicle is a computing device installed in and / or part of the vehicle.   The computing device is part of a wireless network, and the geographic location of each of at least some of the data samples is one or more wireless network access points with which the computing device is communicating. The providing of the multiple data samples in the single data transmission includes transmitting the single data transmission over the wireless network based at least in part on 74. The method according to 74.   The device in the vehicle includes a satellite transmitter, and the providing of the multiple data samples in the single data transmission includes transmitting the single data transmission via the satellite transmitter. 53. The method of claim 52, wherein:   Prior to the provision of the multiple data samples in the single data transmission, changing information associated with the multiple data samples to reduce the amount of data included in the single data transmission. 53. The method of claim 52, further comprising:   Obtaining one or more additional data samples each including information relating to movement of the vehicle prior to the provision of the multiple data samples in the single data transmission; and 53. The method of claim 52, further comprising determining to exclude the additional sample from the multiple data samples provided in transmission.   79. The determination of claim 78, wherein the decision to exclude the additional data sample is based at least in part on determining that the additional data sample does not reflect a movement of interest. The method described.   The use of the data samples provided from the plurality of vehicles is based on at least a portion of one or more roads based at least in part on the reported geographical location and movement characteristics from the provided data samples. Evaluating the road movement conditions of a vehicle moving in the road and the evaluated road to assist the potential movement of other vehicles on or near the at least part of the one or more roads 53. The method of claim 52, comprising using a movement condition.   81. The method of claim 80, wherein the evaluated road movement condition comprises an average speed of all vehicles traveling on the at least part of the one or more roads.   The use of the evaluated road movement condition to assist the other moving vehicle includes information about the evaluated road movement condition in a substantially real-time manner after the evaluation. And enabling information on future road movement conditions that are predicted based at least in part on the evaluated road movement conditions. 81. The method of claim 80.   81. The method of claim 80, wherein the evaluation of the road movement condition and the use of the evaluated road movement condition are performed automatically under control of a traffic information system remote from the vehicle. .   The one or more predefined conditions include a current ability to perform wireless data transmission, and it is temporarily impossible to perform wireless data transmission that the input of the multiple data samples. The method includes obtaining another data sample when wireless data transmission is available, and sending each of the other data samples to the traffic information system without storing the other data sample. 53. The method of claim 52, further comprising providing.   53. The method of claim 52, wherein the one or more predefined conditions include an arrival of the vehicle at a location where the single data transmission can be performed.   The ability to determine the geographical location of the vehicle is temporarily unavailable and of the data samples acquired while the ability to determine the geographical location of the vehicle is temporarily unavailable The at least one respective geographic location of the vehicle is outside of the geographic location from one or more other data samples obtained while the ability to determine the geographic location of the vehicle is available. 53. The method of claim 52, based at least in part on interpolation and / or interpolation.   The vehicle is one of the multiple vehicles each configured to provide data samples to a single traffic information system to reflect the movement of the multiple vehicles. Item 53. The method according to Item 52.   88. The method of claim 87, wherein the multiple vehicles are part of a fleet of vehicles that operate in conjunction with each other.   88. The method of claim 87, wherein the multiple vehicles are instructed to operate in a separate manner to create a variation of the information provided regarding the movement of the multiple vehicles.   53. The method of claim 52, wherein the vehicle is instructed by the traffic information system to move on the route to obtain information regarding vehicle movement along the indicated route. A computer-readable medium, the content of which enables a computing device to provide information regarding movement of the mobile device by performing the method, the method comprising:
Obtaining a number of data samples at a number of times, each data sample indicating a geographical location of the mobile device at the time associated with the data sample;
Storing the multiple data samples;
Automatically providing information from the multiple stored data samples for use in facilitating movement of another mobile device.   The mobile device is associated with a vehicle moving on one or more roads, and at least some of the data samples each have a movement characteristic of the associated vehicle at the time associated with the data sample. 92. The computer readable medium of claim 91, as shown.   The movement characteristics of each of the at least some data samples include a reported direction of travel of the associated vehicle and a reported speed of the associated vehicle, from the multiple stored data samples The provision of the information provides a single data transmission to a traffic information system for use with data samples from a number of other vehicles to facilitate movement of other vehicles on the one or more roads. 94. The computer readable medium of claim 92, comprising transmitting the multiple data samples at.   94. The method of claim 92, wherein the method is performed by one or more devices located along the one or more roads that can detect passage of the mobile device and / or the vehicle. The computer-readable medium described.   The method includes information from the multiple stored data samples to reduce the amount of data included in the single data transmission prior to the provision of the information from the multiple stored data samples. 94. The computer-readable medium of claim 91, wherein the provided information includes the aggregated information.   The mobile device is at least part of the computing device, the mobile device has the ability to access geolocation determination functions, and the geographic location of each of at least some of the data samples 94. The computer readable medium of claim 91, wherein is based on information obtained from the mobile device.   92. The computer readable medium of claim 91, wherein the computer readable medium is a memory of a computing device.   92. The computer readable medium of claim 91, wherein the computer readable medium is a data transmission medium that transmits a generated data signal including the content.   94. The computer-readable medium of claim 91, wherein the content is instructions that when executed cause the computing device to perform the method. One or more storage components;
Obtaining the multiple data samples at the acquisition time such that each of the multiple data samples reflects one or more movement characteristics of a vehicle at multiple separate acquisition times of the data sample; The plurality of data samples for use by a remote traffic information system in temporarily storing the obtained data samples in the one or more storage components and facilitating travel on one or more roads One or more data source information provision components configured to provide information regarding movement of the vehicle on one or more roads by sending stored data samples as a group Characterized in that it is configured to provide information on the movement of vehicles on the road Computing device.   The data source information provision component may store the stored data to allow other stored data samples to be temporarily stored after the transmission as a group of the plurality of stored data samples. Further configured to remove samples from the one or more storage components, the data source information provision component repeatedly obtains and stores data samples and periodically transmits the stored data samples. 101. The computing device of claim 100, further configured to repeatedly perform the providing of the information regarding the movement of the vehicle.   The computing device includes a receiver that travels with the vehicle and that can obtain a Global Positioning System (“GPS”) signal, wherein the mobility characteristics reflected in each of the data samples are determined by the geographic location of the vehicle. The computing device of claim 100, comprising at least one of position, velocity, and direction of travel.   The computing device is part of a system that includes a wireless transmitter capable of transmitting data transmissions, wherein the transmission as a group of the multiple stored data samples is transmitted to other ones on the one or more roads Wireless to transmit the multiple stored data samples in a single data transmission to the remote traffic information system for use with data samples from multiple other vehicles in facilitating vehicle movement 104. The computing device of claim 102, comprising using a transmitter.   The computing device of claim 100, wherein the data source information provision component includes software instructions that are executed in a memory of the computing device.   The data source information provision component is configured such that the multiple data samples at the acquisition time such that each of the multiple data samples reflects one or more movement characteristics of the vehicle at multiple separate acquisition times of the data sample. Remote traffic information in acquiring the multiple acquired data samples in the one or more storage components and facilitating travel on one or more roads A means for providing information regarding movement of the vehicle on one or more roads by transmitting the multiple stored data samples as a group for use by a system. 100. The computing device according to (100). A computer-implemented method for evaluating a data sample reported by a vehicle traveling on a road, the data sample comprising information regarding the movement of the vehicle;
Receiving an indication of a number of road segments of one or more roads;
Receiving information related to the current traffic conditions of the multiple road segments, the received information being reported from one of the multiple vehicles, each of the road segments at the reported time Receiving a plurality of data samples that reflect the reported speed of the one vehicle at a reported location associated with one of them;
For each of the multiple road segments,
Identifying the group such that the data sample of a group of multiple data samples from the plurality of data samples has a reported location corresponding to movement on the road segment;
For each of the data samples in the group, determine the average rate and standard deviation of these data samples based on the reported rate of all other data samples in the group and the reported of the data samples The data sample is a relative statistical outlier relative to the other data samples of the group based on how comparable the difference between the determined speed and the determined average speed is to the determined standard deviation Determining whether or not
Excluding the data sample determined to be a statistical outlier from the group;
After the exclusion, using the data samples remaining in the group to infer traffic conditions of all vehicles traveling on the road segment, so that the data samples are based on the data samples Inferred traffic conditions are made available for use in facilitating travel on the road segment, and based on data samples identified to represent movement on the road segment. Evaluating the traffic conditions of the road segment.   The assessment of the traffic conditions of each of the multiple road segments is performed for each of a number of distinct fixed periods, and the identification of the group of road segment data samples for the fixed period is the identified of the group. 107. The method of claim 106, further executed such that a data sample has a reported time corresponding to the period of time.   For each of at least one of the multiple road segments, one of the data samples of the identified group of the one road segment determined to be a statistical outlier is another road segment. 107. The method of claim 106, wherein the sample is a data sample that is illegally associated with the one road segment from a moving vehicle.   For each of at least one of the plurality of road segments, one of the data samples of the identified group of the one road segment determined to be a statistical outlier is the one road 107. The method of claim 106, wherein the sample is a data sample from a vehicle parked on or beside the segment.   The evaluation of the traffic situation of each of the multiple road segments determines all average speeds and standard deviations of the data samples of the identified group of the road segment; and Using all the determined average velocities and standard deviations of the data samples as part of the determination of the average velocities and standard deviations of all the other data samples of the group for each of the data samples; 107. The method of claim 106, comprising:   The determination of whether a data sample of a group is a statistical outlier relative to the other data sample of the group is determined by determining the reported rate of the data sample and the other data sample of the group. 111. The method of claim 110, comprising determining whether the difference from a determined average velocity exceeds a threshold based on the determined standard deviation of the other data samples of the group. Method.   Data samples relating to the current traffic conditions of the multiple road segments are repeatedly received to reflect changing traffic conditions, and the evaluation of the traffic conditions of each of the multiple road segments is in real time form. 107. The method of claim 106, wherein the method is performed on a recently received data sample.   The use of data samples remaining in a group for inferring traffic conditions for all vehicles traveling on a road segment includes determining an average speed of the remaining data samples and determining the determined average Inferring an average speed of all vehicles traveling on the road segment based on speed, and information about the inferred average speed to one or more persons considering upcoming movement on the road segment 113. The method of claim 112, comprising providing. A computer-implemented method for evaluating a data sample representing a vehicle traveling on a road, comprising:
Receiving an indication of one or more segments of one or more roads, each road segment having a number of associated data samples each reflecting a reported speed of a vehicle on the road segment. Having, receiving,
For each of at least one of the road segments,
Automatically analyzing the multiple related data samples of the road segment to determine one or more of the data samples that do not represent actual vehicle movement on the road segment, comprising: Automatically analyzing that at least one of the determined data samples is a statistical outlier with respect to other data samples of the plurality of related data samples;
One or more for excluding the determined data sample from later use in order to make the other data sample available for use when facilitating travel on the road segment. Providing a display.   For the one or more of the at least one road segment, the provision of the display for excluding the determined data sample from later use determines an average speed of a vehicle traveling on the road segment. Analyzing the associated data sample of the road segment other than the determined data sample to determine the use for determination when facilitating movement of other vehicles on the road segment. 115. The method of claim 114, further comprising indicating an average speed.   For the one or more of the at least one road segment, the provision of the display for excluding the determined data sample from subsequent use determines a traffic flow of a vehicle traveling on the road segment. Analyzing the associated data sample of the road segment other than the determined data sample to determine the determination for use in facilitating movement of other vehicles on the road segment. 115. The method of claim 114, further comprising indicating a traffic flow.   For one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is determined by 115. The method of claim 114, comprising determining that each is a statistical outlier with respect to other data samples of the multiple data samples associated with the road segment.   The determination that a data sample associated with a road segment is a statistical outlier with respect to other data samples of the large number of data samples associated with the road segment uses only one outlier analysis. 118. The method of claim 117, wherein the method is performed.   The determination that a data sample associated with a road segment is a statistical outlier with respect to other data samples of the multiple data samples associated with the road segment is reflected by each of the data samples 118. The method of claim 117, wherein the method is based at least in part on the speed to be achieved. The determination that each of the one or more data samples of a road segment is a statistical outlier with respect to other data samples of the multiple data samples associated with the road segment is:
Determining all mean speeds and standard deviations of the multiple data samples of the road segment;
For each of the one or more data samples of the road segment,
Determining an average speed and standard deviation of all other data samples of the multiple data samples of the road segment based on the determined average speed and standard deviation of all of the multiple data samples of the road segment; And
Determining a difference between the reported speed of the data sample and the determined average speed of all other data samples of the multiple data samples of the road segment;
Determining a threshold based at least in part on the determined standard deviation of all other data samples of the multiple data samples of the road segment;
118. The method of claim 117, comprising: identifying the data sample as a statistical outlier when the determined difference exceeds the determined threshold.   For each of the one or more road segments, each of the one or more data samples of the road segment is a statistical outlier with respect to other data samples of the multiple data samples associated with the road segment. 118. The method of claim 117, wherein the determination of being is performed in a substantially real-time manner.   For one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is determined by its reported speed Assessing the activity of each vehicle reflected by at least one of the determined data samples, and each of the assessed activities of the vehicle does not correspond to actual vehicle movement on the road segment 115. The method of claim 114, comprising determining.   The method of claim 122, wherein the assessed activity of at least one of the vehicles corresponds to a parked vehicle.   The assessed activity of at least one of the vehicles based on one or more determined data samples is a movement on a road segment other than the road segment to which the one or more determined data samples are associated. 122. The method of claim 122, corresponding to:   The determination of one or more data samples of the road segment that does not represent actual vehicle movement on the road segment for one or more of the at least one road segment is moved over the road segment Identifying a number of data samples reported by a single vehicle, determining activity of the single vehicle over time based on the identified data samples, and determining the determined activity 115. The method of claim 114, comprising: based on determining that the identified data sample does not represent actual vehicle movement on the road segment.   The determination of one or more data samples of the road segment that does not represent actual vehicle movement on the road segment for one or more of the at least one road segment is associated with the road segment 115. Identifying the expected value of the multiple data samples and determining that the determined data sample does not apply to the identified expected value. The method described in 1.   The determination of one or more data samples of the road segment that does not represent actual vehicle movement on the road segment for one or more of the at least one road segment is associated with the road segment 115. Determining a statistical distribution of the multiple data samples and determining that the determined data samples do not fit into the determined statistical distribution. the method of.   For one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is obtained by determining a number of road segments Identifying distinct data curves, each of said data curves reflecting a distinct subset of vehicle movement over at least a portion of said road segment, and said determined data sample is said data curve 115. The method of claim 114, wherein the method applies to at least one data curve that reflects a subset of vehicle movements that are not of interest.   129. The method of claim 128, wherein at least one of the identified data curves is a Gaussian curve.   For one or more of the at least one road segment, the multiple related data samples of the road segment each have a reported time corresponding to the reported speed of the vehicle of the data sample. Reflecting further, the automatic analysis of the multiple related data samples of the road segment may be performed during the predetermined period of time such that the actual vehicle movement on the road segment is a movement during a predetermined period of time. 115. The method of claim 114, further corresponding.   For each of the one or more road segments, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is determined by each of the determined data samples. 131. The method of claim 130, comprising identifying that the reported time of is not within the predetermined period of time for the road segment.   For each of a number of distinct time periods, a number of associations of the one road segment each reflecting a reported speed of the vehicle at a reported time on one of the road segments during the time period The automatic analysis is performed for the one road segment for each of the time periods based on the data samples whose reported time is during the time period. 119. The method of claim 114, characterized in that   The determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment for one or more of the at least one road segment is in a substantially real-time manner. 115. The method of claim 114, wherein the method is performed at.   At least some of the multiple data samples associated with at least some of the road segments are acquired by and reported from vehicles traveling on the road segments, and the at least some data 134. The method of claim 133, wherein the reporting of samples occurs in a substantially real-time manner after acquisition of one or more of the at least one data sample. A computer readable medium, the contents of which enable a computing device to evaluate a data sample representing a moving vehicle by performing the method, the method comprising:
Receiving an indication of a number of data samples, each of which reflects the reported travel characteristics of one of a number of vehicles traveling on the road;
Automatically determining one or more of the multiple data samples that do not represent actual vehicle movement on the road;
Providing one or more representations of the data sample other than the determined data sample to enable the displayed data sample to be used for facilitating travel on the road A computer-readable medium comprising:   The reported movement characteristics of each data sample include the reported speed of the vehicle to which the data sample corresponds, and the one or more data samples that do not represent actual vehicle movement on the road At least one of the determinations includes determining that the at least one data sample is a statistical outlier with respect to other data samples of the data sample, wherein the providing of the indication of the data sample comprises: Analyzing the data samples to determine an average speed of a vehicle traveling on the road and determining the average speed for use in facilitating movement of other vehicles on the road 136. The computer-readable medium of claim 135, including:   The determination that the at least one data sample is a statistical outlier with respect to other data samples of the data sample includes performing only one excluded outlier analysis in a substantially real-time manner. 136. The computer readable medium of claim 136.   136. The computer readable medium of claim 135, wherein the computer readable medium is a memory of a computing device.   136. The computer readable medium of claim 135, wherein the computer readable medium is a data transmission medium that transmits a generated data signal including the content.   136. The computer-readable medium of claim 135, wherein the content is instructions that when executed cause the computing device to perform the method. A computing system configured to evaluate a data sample representing a moving vehicle, comprising:
A first component configured to receive, for each of a number of roads, an indication of a number of data samples of the road, each reflecting a reported speed of a vehicle traveling on the road;
For each of the multiple roads,
Automatically determining one or more of the multiple data samples of the road that are statistically abnormal with respect to other data samples of the multiple data samples of the road;
One or more of the multiple data samples of the road other than the determined data sample so that the data sample shown can be used for use in facilitating travel on the road And a data sample outlier eliminator component configured to provide an indication.   For each of at least one of the multiple roads, the determination of the data sample of the road that is a statistical outlier with respect to other data samples of the road is substantially in real-time form. The providing of the display of the data sample of the road analyzes the data sample to determine an average speed of a vehicle traveling on the road. 142. The computing system of claim 141, including: and indicating the determined average speed for use in facilitating movement of other vehicles on the road.   142. The computing system of claim 141, wherein the first component and the data sample outlier eliminator component each include software instructions that are executed in a memory of the computing system.   The first component comprises means for receiving, for each of a number of roads, an indication of a number of data samples of the road, each reflecting a reported speed of a vehicle traveling on the road, The eliminator component may, for each of the multiple roads, determine one or more of the multiple data samples of the road that are statistically abnormal with respect to other data samples of the multiple data samples of the road. Of the multiple data samples of the road other than the determined data samples to automatically determine and make the indicated data samples available for use in facilitating travel on the road 142. The method of claim 141, comprising means for providing one or more displays. Placing the computing system. A computer-implemented method for determining estimated traffic flow information for a road based on a data sample reported by a vehicle traveling on a road, including information relating to the movement of the vehicle,
Receiving an indication of a number of road segments of one or more roads;
For each of a number of distinct observed quantities of vehicles on a road segment over a period of time, a probability distribution indicating the probability of the vehicle's arrival rate amount to the road segment given the observed quantity of vehicles. Generating step;
For each of a number of fixed periods,
Receiving information relating to current traffic conditions of the multiple road segments during the period of time, each of the received information being reported from one of a number of vehicles, during the period of time; A plurality of data samples for the period of time reflecting the reported speed of the one vehicle at a reported location on one of the road segments at a reported time of Vehicles are a subset of all vehicles traveling on the road segment during the period of time, and the received information is from one of a number of traffic sensors each monitoring the multiple road segments. One of the one or more vehicles at a location above one of the road segments at one or more of the reported times during the period of time reported Others reflect the rate reported based on reading of a plurality of speeds, including a plurality of additional data samples of said predetermined period, comprising: receiving,
For each of the multiple road segments,
Identifying a group of multiple data samples of the road segment for the period of time, the multiple data samples from at least one of the plurality of data samples and the plurality of additional data samples. To identify,
Determining the amount of vehicles to which the data samples of the group correspond, wherein the corresponding vehicles are a subset of all the vehicles traveling on the road segment during the period of time. When,
Determining the most probable traffic arrival rate amount to the road segment of all the vehicles traveling on the road segment during the predetermined period, wherein the determination of the traffic arrival rate amount comprises: Determining based at least in part on the generated probability distribution of the determined amount of vehicles reporting the data samples of the group;
Determining the most probable traffic density of the road segment to represent the total amount per unit of distance of all the vehicles traveling on the road segment during the fixed period, wherein the determination of the traffic density is Determining based at least in part on the determined traffic arrival rate amount of the road segment during the period of time;
Determining the most probable percentage traffic occupancy of at least one point on the road segment by all the vehicles moving on the road segment during the period of time, wherein the determination of the percentage traffic occupancy is Determining based at least in part on the determined traffic density;
Automatically estimating traffic flow information of all the vehicles traveling on the road segment during the period of time,
Using at least some of the determined traffic arrival rate amount, traffic density, and percentage traffic occupancy for the period of time to facilitate future travel on the one or more roads;
As a result, the estimated traffic flow information is determined for the road segment based on data samples reflecting actual vehicle movement on the road segment. Method.   For each of at least one of the time periods and at least one of the road segments, the identified group of a number of data samples of the road segments of the time period has a reported position of the Including a number of data samples from the plurality of data samples of the time period on the road segment, the position of the plurality of data samples from the plurality of additional data samples of the time period being on the road segment. The estimation of the traffic flow information of all the vehicles traveling on the road segment during the fixed period is based on the fixed period based at least in part on the reported speed of the data samples of the group All of the above moving on the road segment The method of claim 145, characterized in that it comprises generating a mean traffic rate that is both estimated.   For each of the at least one fixed period and each of the at least one road segment, the determination of the traffic density of the road segments of all the vehicles traveling on the road segment during the fixed period further includes: 147. The method of claim 146, based at least in part on the estimated average traffic speed of the all vehicles traveling on the road segment during the period of time.   The percentage traffic occupancy of at least one point on the road segment by all the vehicles moving on the road segment during the fixed period for each of the at least one fixed period and for each of the at least one road segment The determination further comprises estimating the estimated average traffic speed of all the vehicles traveling on the road segment during the fixed period and the estimation of all the vehicles moving on the road segment during the fixed period. 148. The method of claim 147, based at least in part on the measured average length.   For each of the at least one fixed period and each of the at least one road segment, the determination of the amount of traffic arrival rate to the road segment of all the vehicles moving on the road segment during the fixed period is The data sample of the group is further based at least in part on an estimated percentage of all the vehicles that move on the road segment during the period of time that is the corresponding vehicle. 148. The method according to 148.   For each of at least one of the time periods and at least one of the road segments, the determination of the traffic arrival rate amount of all the vehicles traveling on the road segment during the time period is: Generating a confidence value for the traffic arrival rate quantity to reflect a possible degree of error in the traffic arrival rate quantity, and wherein the at least some determined traffic arrival rate quantities for the period of time 146. The method of claim 145, wherein use includes using the generated confidence value to facilitate the future movement.   The estimation of the traffic flow information of all the vehicles traveling on a road segment during a period is performed in real time using recently received data samples, and the at least some of the period The use of the determined traffic arrival rate amount, the determined traffic density, and the determined percentage traffic occupancy is further substantially facilitated for impending future movement on the one or more roads. 145. The method of claim 145, wherein the method is performed in real time.   The use of the at least some determined traffic arrival rate amount, traffic density, and percentage traffic occupancy for a period of time to one or more persons considering upcoming travel on the one or more roads 153. The method of claim 151, comprising providing information regarding the determined traffic arrival rate amount, traffic density, and percentage traffic occupancy. A computer-implemented method for determining estimated traffic flow information for vehicles traveling on these roads based on data samples reflecting movement on the roads, comprising:
Receiving an indication of one or more segments of one or more roads, each road segment reflecting a movement on the road segment by one of a number of vehicles at a reported time; Receiving a number of associated data samples, each of which is reported by the one vehicle for receiving,
For each of at least one of the road segments,
Identifying the group of multiple data samples associated with the road segment whose reported time occurs during a period of time;
Determining the amount of vehicles reporting the data samples of the group, wherein the vehicles reporting the data samples of the group are a subset of all vehicles that have traveled on the road segment during the period of time. To determine,
Probabilistically evaluate an estimated total amount of all the vehicles that have traveled on the road segment during the period of time based at least in part on the determined amount of the vehicle reporting the data sample. Automatically estimating the traffic flow of a vehicle traveling on the road segment during the period of time, and
Using one or more of the estimated total amount of vehicles to facilitate travel on the one or more roads.   For one or more of the at least one road segment, the probabilistic evaluation of the estimated total amount of all the vehicles that have traveled on the road segment during the period is determined during the period. 154. The method of claim 153, comprising determining a most probable total amount of all the vehicles that have traveled on the road segment.   For one or more of the at least one road segment, the probabilistic evaluation of the estimated total quantity of all the vehicles that have traveled on the road segment during the fixed period is the estimated total quantity 154. The method of claim 153, comprising determining a confidence value for the estimated total amount based at least in part on the likelihood of.   For one or more of the at least one road segment, the probabilistic evaluation of the estimated total amount of all the vehicles that have traveled on the road segment during the period is Determining a traffic arrival rate amount of the vehicle to at least one point of the road segment, the determined traffic arrival rate amount of all the vehicles that have traveled on the road segment during the period of time. 154. The method of claim 153, based at least in part on the estimated total amount.   For each of the one or more road segments, the determination of the traffic arrival rate amount of a vehicle to at least one point of the road segment during the period of time is based on the data samples of the group of the road segments. 157. The method of claim 156, comprising using a probability distribution given the determined amount of reported vehicles and representing the probability of the actual arrival rate amount of the vehicle.   For each of the one or more road segments, the use of the probability distribution as part of the determination of the traffic arrival rate amount of the vehicle to at least one point of the road segment during the period of time is 158. The method of claim 157, comprising determining a most probable traffic arrival rate amount.   For each of the one or more road segments, the use of the probability distribution as part of the determination of the traffic arrival rate amount of the vehicle to at least one point of the road segment during the period of time is 158. The method of claim 157, comprising estimating a reliability of the determined traffic arrival rate amount based on the probability distribution.   158. The method of claim 157, wherein for each of the one or more road segments, the probability distribution is a Poisson distribution.   For each of the one or more road segments, the determination of the traffic arrival rate amount of the vehicle to at least one point of the road segment during the certain period is determined by the control rate of the road segment during the certain period The dominance rate is an estimated percentage of all the vehicles that have traveled on the road segment during the period of time that are vehicles that have reported the data samples of the group of the road segment. 156. The method of claim 156, wherein:   156. For each of the one or more road segments, the at least one point of the road segment from which the traffic arrival rate amount of a vehicle is determined is the beginning of the road segment. The method described.   156. For each of the one or more road segments, the at least one point of the road segment from which the traffic arrival rate amount of a vehicle is determined is all of the road segments. The method described.   For one or more of the at least one road segment, the probabilistic evaluation of the estimated total amount of all the vehicles that have traveled on the road segment during the period is Determining the traffic density of the road segment, wherein the determined traffic density is based at least in part on the estimated total amount of all the vehicles that have traveled on the road segment during the period of time. 154. The method of claim 153.   For each of the one or more road segments, the determined traffic density represents a total amount per unit of distance of all the vehicles that have moved on the road segment during the certain period. 164. The method of claim 164.   For each of the one or more road segments, the determination of the traffic density of the road segment is based at least in part on at least one determined traffic arrival rate amount of the road segment during the period of time. 166. The method of claim 164, wherein:   For each of the one or more road segments, the at least one traffic arrival rate amount of the road segment during the time period to represent a vehicle arriving at at least one point of the road segment during the time period 171. The method of claim 166, further comprising determining.   For each of the one or more road segments, the determination of the traffic density of the road segment is further based on an estimated average traffic speed of all the vehicles that have traveled on the road segment during the period of time. 173. The method of claim 166, based at least in part.   169. The method of claim 168, further comprising estimating, for each of the one or more road segments, the average traffic speed of all the vehicles that have traveled on the road segment during the period of time. the method of.   166. The method of claim 164, for each of the one or more road segments, the determination of the traffic density of the road segment includes determining a most probable traffic density.   165. For each of the one or more road segments, the determination of the traffic density of the road segment includes estimating a reliability of the determined traffic density. Method.   For one or more of the at least one road segment, the probabilistic evaluation of the estimated total amount of all the vehicles that have traveled on the road segment during the period is Determining the traffic occupancy of at least one point on the road segment, wherein the determined traffic occupancy is in the estimated total amount of all the vehicles that have traveled on the road segment during the period of time. 154. The method of claim 153, based at least in part.   For each of the one or more road segments, the determined traffic occupancy of the at least one point on the road segment is that the at least one point has moved over the road segment during the period of time. 173. The method of claim 172, representing an average percentage of time occupied by at least one of the vehicles.   For each of the one or more road segments, the determination of the traffic occupancy of the road segment includes determining at least one determined traffic density of the road segment during the period and the road segment during the period. 178. The method of claim 172, based at least in part on at least one estimated length of the vehicle that has traveled thereon.   For each of the one or more road segments, the determination of the traffic occupancy of the road segment is further at least partially in at least one determined traffic arrival rate amount of the road segment during the period of time. 175. The method of claim 174, wherein:   For each of the one or more road segments, the determination of the traffic occupancy of the road segment is further based on an estimated average traffic speed of all the vehicles that have traveled on the road segment during the period of time. 178. The method of claim 175, based at least in part.   Determining, for each of the one or more road segments, the at least one traffic density of the road segment during the period of time, and the at least one traffic arrival rate amount of the road segment during the period of time. 177. The method of claim 176, further comprising: determining an average traffic speed of all the vehicles that have traveled on the road segment during the period of time.   173. The method of claim 172, for each of the one or more road segments, the determination of the traffic occupancy of the road segment includes determining a most probable traffic occupancy.   173. The method of claim 172, for each of the one or more road segments, the determination of the traffic occupancy of the road segment includes estimating a reliability of the determined traffic occupancy. Method.   For each one or more of the at least some road segments, the estimation of the traffic flow of a vehicle traveling on the road segment during the period is performed for each of a number of distinct periods. 154. The method of claim 153, wherein:   For each of one or more of the at least some road segments, the estimate of the traffic flow of a vehicle traveling on the road segment during the period of time is calculated from the associated data sample of the road segment. 154. The method of claim 153, wherein the method is performed for each of a number of the time windows such that at least some of them are used for each of a number of overlapping time windows during the period of time. .   6. For each of the one or more road segments, the multiple overlapping time windows during the period of time are changed to reflect one or more current conditions. 181. The method according to 181.   For each one or more of the at least some road segments, the estimate of the total quantity of all the vehicles that have traveled on the road segment during the fixed period is at least one of the estimated total quantities. The use of the one or more estimated total quantities of the vehicle is determined to facilitate future travel on the one or more roads. 154. The method of claim 153, comprising using one or more of the confidence values.   The use of the one or more estimated total quantities of the vehicle allows the person to the one of the vehicles to facilitate a determination by one or more persons regarding travel on the one or more roads. 154. The method of claim 153, comprising providing an indication of a plurality of estimated total quantities.   The evaluation of the one or more estimated total quantities and the use of the one or more estimated total quantities are used in the evaluation to allow a substantially real-time judgment by the person. 185. The method of claim 184, wherein the method is performed in a substantially real-time manner relative to receipt of the data sample. A computer-readable medium, the contents of which enable a computing device to estimate traffic flow information of a moving vehicle by performing the method,
Receiving an indication of a number of data samples each reflecting a movement of one of a number of vehicles on the road;
Probabilistically estimating the traffic flow of all vehicles traveling on the road during a period of time based at least in part on the amount of vehicles reflected by the data samples;
Providing a representation of the estimated traffic flow for use in facilitating travel on the road.   The multiple data samples are each reported by a vehicle, each having information regarding the movement of the vehicle on the road at the reported time during the period of time, and for all vehicles traveling on the road. The estimation of the traffic flow includes determining an amount of vehicles that have reported the data sample, wherein the vehicle that has reported the data sample is the number of all the vehicles that travel on the road during the period of time. And the estimate of the traffic flow of all vehicles traveling on the road is a subset of the vehicles of all the vehicles based at least in part on the determined amount of the vehicles reporting the data sample. Determining an estimated arrival rate to a point on the road and a likelihood of the estimated arrival rate; The computer-readable medium of claim 186 that.   The estimation of the traffic flow of all vehicles traveling on the road further comprises determining an average percentage occupancy of points on the road by the vehicle traveling on the road during the period of time; The determination of average percentage occupancy is based at least in part on the determined estimated arrival rate and the indication of the estimated traffic flow for use in facilitating travel on the road. 189. The computer of claim 187, wherein providing includes presenting information regarding the estimated traffic flow to a vehicle driver for use in influencing decisions regarding travel on the road. A readable medium.   187. The computer readable medium of claim 186, wherein the computer readable medium is a memory of a computing device.   187. The computer readable medium of claim 186, wherein the computer readable medium is a data transmission medium that transmits a generated data signal including the content.   187. The computer readable medium of claim 186, wherein the content is instructions that when executed cause the computing device to perform the method. A computing system configured to estimate traffic flow information of a moving vehicle,
A first component configured to receive, for each of a number of roads, an indication of a number of data samples associated with the road, each including information representing movement of a vehicle on the road;
For each of the multiple roads,
Determining the amount of vehicle whose movement is represented by the information of the data sample associated with the road;
Generating a stochastic estimate of the traffic flow of all vehicles traveling on the road during a period of time based at least in part on the determined amount of vehicles;
And a data sample flow assessor component configured to provide an indication of the estimated traffic flow for use in facilitating travel on the road.   For each of at least some of the multiple roads, the multiple data samples associated with the road are such that the information contained in the data samples is reported at the time of the road during the period of time. The vehicle, each reported by the vehicle to be information relating to the movement of the vehicle on a road, and whose movement is represented by the information in the data sample associated with the road, during the certain period of the road The generation of the probabilistic estimate of the traffic flow of all the vehicles traveling on the road is a subset of all the vehicles traveling on the road All based at least in part on the determined amount of the vehicle represented by the information in the data sample Determining the estimated arrival rate of the vehicle to a point on the road and the likelihood of the estimated arrival rate, the estimated for use in facilitating movement on the road Providing said indication of traffic flow comprising presenting information relating to said estimated traffic flow to a vehicle driver for use in influencing judgments relating to travel on said road 192. The computing system of claim 192.   For each of the at least some roads, the generation of the probabilistic estimate of the traffic flow of all the vehicles traveling on the road is the road moving on the road during the period of time. 193. The method includes determining an average percentage occupancy of points on the road by a vehicle, wherein the determination of the average percentage occupancy is based at least in part on the determined estimated arrival rate. The computing system as described in.   193. The computing system of claim 192, wherein the first component and the data sample flow assessor component each include software instructions that are executed in a memory of the computing system.   The first component comprises means for receiving, for each of a number of roads, a display of a number of data samples associated with the road, each of which includes information representative of vehicle movement on the road, the data sample flow assessor component Determining, for each of the plurality of roads, a quantity of vehicles whose movement is represented by the information in the data sample associated with the roads and is constant based at least in part on the determined quantity of vehicles Probabilistic estimates of the traffic flow of all vehicles traveling on the road during the period and providing an indication of the estimated traffic flow for use in facilitating travel on the road 193. The computing system of claim 192, comprising means. A computer-implemented method for evaluating a data sample reported by a vehicle traveling on a road, wherein the data sample includes information regarding the movement of the vehicle, the method comprising:
Receiving an indication of a number of road segments of one or more roads;
Receiving information related to current traffic conditions of the multiple road segments, wherein the received information is reported from one of a number of vehicles, each at the reported geographical location. Receiving a plurality of data samples reflecting the reported speed of one vehicle and reflecting the reported direction of travel of the one vehicle;
For each of the multiple road segments,
Identifying a group of multiple data samples from the plurality of data samples, wherein the data samples of the group are within a predefined distance from one or more predefined geographic locations of the road segment. Identifying to have a reported geographical location and having a reported travel direction within a predefined difference from one or more predefined travel directions of the road segment;
Based at least in part on the reported geographical location of the determined road segment does not correspond to a predefined portion of the road segment for which vehicle movement is of interest. Automatically determining that one or more of the data samples do not represent actual vehicle movement on the road segment;
Excluding from the group the data samples determined not to represent actual vehicle movement on the road segment;
After the exclusion is identified as representing movement on the road segment by using the data samples remaining in the group to infer traffic conditions of all vehicles traveling on the road segment. Evaluating the traffic situation of the road segment based on the data sample, so that the traffic situation inferred based on the data sample is used for use in facilitating movement on the road segment. A method characterized by comprising the step of assessing.   One of the multiple road segments corresponds to a first section of a highway having multiple lanes, and the predefined geographic location of the one road segment is the highway's relative to the first section. The predefined distance includes a geographic area that covers the multiple lanes and the geographic area is extended by at least some of the data samples of the identified group of the one road segment. Based at least in part on the degree of accuracy of a location device of the type used to determine the reported geographical location, the predefined direction of travel of the one road segment is the One or more travel directions corresponding to the direction of vehicle movement on the multiple lanes of the highway with respect to the first section The predefined difference in the one or more predefined travel directions of the one road segment is at least some of the data samples of the identified group of the one road segment 197. The method of claim 197, wherein the method is based at least in part on the degree of accuracy of a direction determination device of the type used to determine the reported direction of travel.   The predefined portion of the one road segment in which vehicle movement is of interest includes one or more lanes of the highway and does not represent actual vehicle movement on the one road segment One of the data samples of the identified group of the one road segment determined to be determined that its reported geographic location corresponds to a ramp to and / or from the highway. 199. The method of claim 198, wherein the method is a data sample.   The predefined portion of the one road segment in which vehicle movement is of interest includes one or more lanes of the highway and does not represent actual vehicle movement on the one road segment One of the data samples of the identified group of the one road segment determined to be corresponding to a collector / distributor road or collector / distributor lane whose reported geographical location is near the highway 199. The method of claim 198, wherein the data sample is determined to be.   The predefined portion of the one road segment in which vehicle movement is of interest includes a subset of the highway lanes and is determined not to represent actual vehicle movement on the one road segment. One of the data samples of the identified group of the one road segment is data determined to correspond to a lane of the highway whose reported geographical location is not included in the subset of lanes 199. The method of claim 198, wherein the method is a sample.   One of the data samples of the identified group of the one road segment determined not to represent actual vehicle movement on the one road segment is in a separate second portion of the highway. 199. The method of claim 198, wherein the data sample is determined to be fraudulently associated with the one road segment from a vehicle traveling on another corresponding road segment.   The type of position determination device is a type of global positioning system (GPS) device, and the predefined distance by which the geographic area is extended can be read from the GPS device of that type. 199. The method of claim 198, corresponding to a distance that is accurate.   For each of one or more of the multiple road segments, the automatic determination that the one or more data samples of the group do not represent actual vehicle movement on the road segment further includes: Based on at least in part the one or more movement characteristics of the vehicle that reported these data samples, other than the reported geographical location, the one or more movement characteristics are determined by the reporting of these data samples. 197. The method of claim 197, comprising a speed of the vehicle as reflected by a speed being measured.   Each of the plurality of data samples further indicates a reported time associated with the reported speed and reported geographical location and reported travel direction of the data sample, and each of the plurality of road segments The evaluation of the traffic situation is performed for each of a number of distinct fixed periods, and the identification of the group of road segment data samples for a fixed period is determined when the identified data sample of the group is the fixed period of time. 199. The method of claim 197, further executed to have a reported time corresponding to.   Data samples relating to the current traffic conditions of the multiple road segments are repeatedly received to reflect changing traffic conditions, and the evaluation of the traffic conditions of each of the multiple road segments is in real time form. The use of data samples remaining in the group for inferring traffic conditions of all vehicles traveling on a road segment, performed on recently received data samples at the average of the remaining data samples Determining a speed, inferring an average speed of all vehicles traveling on the road segment based on the determined average speed, and information about the inferred average speed coming on the road segment Providing to one or more persons considering power movements The method according to 97. A computer-implemented method for evaluating a data sample representing a vehicle traveling on a road, comprising:
Receiving the indication of one or more segments of one or more roads, each road segment corresponding to the road segment, each reported by one of a number of vehicles Receiving, having a number of associated data samples indicating a reported position of
For each of at least one of the road segments,
Automatically analyzing the multiple related data samples of the road segment to determine one or more of the data samples that do not represent actual vehicle movement on the road segment, comprising: At least one of the determined data samples is the reported location of the vehicle that reported the data sample, respectively, such that the reported location does not correspond to the actual vehicle movement on the road segment. Wherein at least one of the determined data samples is such that an associated direction of travel of the vehicle reporting the data sample does not correspond to the actual vehicle movement on the road segment. Automatically analyzing with an associated direction of travel;
One or more indications for excluding the determined data sample from later use so that other data samples are available for use in facilitating travel on the road segment. Providing a computer-implemented method.   For each of one or more of the at least one road segment, the provision of the display for excluding the determined data sample from later use is an average speed of a vehicle traveling on the road segment Analyzing the associated data sample of the road segment other than the determined data sample to determine and determining for use in facilitating movement of other vehicles on the road segment 207. The method of claim 207, comprising: indicating a measured average speed.   For each of the one or more road segments, the provision of the display for excluding the determined data sample from subsequent use is for determining traffic flow of a vehicle traveling on the road segment Analyzing the associated data sample of the road segment other than the determined data sample and determining the determined traffic flow for use in facilitating movement of other vehicles on the road segment. 209. The method of claim 208, further comprising: indicating.   For each of one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is based on these data Determining that the reported vehicle position of a sample corresponds to a section of the road that is not of interest for use in representing the actual vehicle movement on the road segment. 207. The method according to item 207.   For each of at least one of the one or more road segments, the section of the road is at least a portion of a smaller capacity road leading to and / or from the road segment. 213. The method of claim 210.   211. For each of at least one of the one or more road segments, the section of road is part of a separate road that is close to the road of the road segment. Method.   211. The method of claim 210, for each of at least one of the one or more road segments, the section of road is a subset of a number of lanes that are part of the road section. .   For each of at least one of the one or more road segments, the section of road includes a ramp to and / or from the road segment, a collector / distributor road associated with the road of the road segment One of the road segment collector / distributor lane, the road segment feeder lane, the road segment shoulder of the road segment, and the road segment collapse area of the road segment. 213. The method of claim 210, wherein the method is at least part of one or more.   213. The method of claim 210, wherein the section of roads that are not of interest for one of the one or more road segments is part of the one road segment.   For one of the one or more road segments, the section of the road that is not of interest is at least part of another road segment distinct from the one road segment, 213. The method of claim 210.   For one of the at least one road segment, at least some of the multiple data samples associated with the one road segment are also associated with one or more other distinct road segments; 207. The method of claim 207, wherein the determined one or more data samples of the one road segment are from the at least some data samples.   For one of the road segments, the multiple related data samples of the one road segment are not interested based at least in part on the one road segment that is not interested. 207. The method of claim 207, further comprising: automatically determining and providing one or more displays to exclude the multiple related data samples from later use. Method.   219. The method of clause 218, wherein the one road segment is determined to be uninterested based at least in part on having a functional road type that is not of interest.   219. The one-way segment of claim 218, wherein the one road segment is determined not to be of interest based at least in part on a determination of an actual amount of vehicle traffic on the one road segment. Method.   The one road segment is at least partially in determining the amount of intraday variability of vehicle traffic on the one road segment and / or determining the amount of daily variability of vehicle traffic on the one road segment. 219. The method of claim 218, wherein the method is determined not to be of interest.   218. The one road segment is determined to be uninterested based at least in part on determining an actual amount of traffic jam on the one road segment. Method.   The one road segment is at least partly for determining the amount of intraday variability of traffic jams on the one road segment and / or for determining the amount of daily variability of traffic jams on the one road segment. 219. The method of claim 218, wherein the method is determined not to be of interest.   To automatically determine that the one or more data samples are not associated with any of the uninteresting road segments, and to exclude the one or more data samples from later use. 207. The method of claim 207, further comprising providing one or more displays.   For each one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is the one 206. The method of claim 207, comprising determining that the one or more data samples do not represent based at least in part on the reported location of a plurality of data samples.   For each of the one or more road segments, the multiple related data samples of the road segment each indicate a speed of the vehicle that reported the data sample, and the one or more of the road segments 226. The method of claim 225, wherein the determination that a data sample does not represent is further based at least in part on the rate indicated by the one or more data samples.   For each of at least one of the one or more road segments, reported by the vehicle that reported the data sample for each of at least some of the multiple related data samples of the road segment. 226. The method of claim 226, further comprising estimating the indicated velocity of the data sample by using a reported location indicated by multiple data samples.   For each of the one or more road segments, the plurality of associated data samples of the road segment each have an associated direction of travel of the vehicle that reported the data sample, and the one of the road segments 226. The method of claim 225, wherein the determination that one or more data samples do not represent is further based at least in part on the direction of travel associated with the one or more data samples. .   For each of at least one of the one or more road segments, reported by the vehicle that reported the data sample for each of at least some of the multiple related data samples of the road segment. 229. The method of claim 228, further comprising estimating the direction of travel associated with the data sample by using a reported location indicated by multiple data samples.   One of the one or more road segments is a part of a road that includes vehicles moving in two opposite directions, and the one road segment moves in one of the two directions The determination that the one or more data samples of the one road segment that corresponds to a vehicle and is at least partially based on the direction of travel associated with the one or more data samples does not 229. The method of claim 228, comprising determining that a traveling direction to travel does not represent a data sample for the one road segment corresponding to movement to the other of the two directions.   One of the one or more road segments is a part of a road including a number of lanes with vehicles moving in a number of directions, the one road segment being a One or more of the one road segment corresponding to a subset of the multiple lanes having vehicles moving to one or more and based at least in part on the direction of travel associated with the one or more data samples The determination that the data sample does not represent comprises determining that a data sample whose associated direction of travel does not correspond to the one or more directions does not represent the one road segment. 229. The method of claim 228, characterized in that   One of the one or more road segments overlaps one or more other road segments of one or more other roads, and vehicles on the one road segment are the other Moving in one or more directions distinct from one or more other directions in which the vehicle on the road segment travels and at least partially in the direction of travel associated with the one or more data samples The determination that one or more data samples of the one road segment based does not represent that the data sample whose associated traveling direction does not correspond to the one or more directions of the one road segment 229. The method of claim 228, comprising determining not to represent one road segment.   For each one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is the one 207. The method of claim 207, comprising comparing a plurality of data samples with at least some of the other data samples of the road segment.   For each one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is of interest. Identifying a subset of the actual vehicle movement on the road segment that is or is not interested and determining whether the one or more data samples correspond to the identified subset 207. The method of claim 207, comprising: performing.   Receiving one or more indications of a plurality of data samples each indicating a reported position of the vehicle; and one or more of the reported positions of the data samples associated with each of the at least one road segment Further comprising associating each of at least some of the plurality of data samples with at least one of the road segments based at least in part on corresponding to a plurality of locations. 207. The method of claim 207.   Each of the plurality of data samples has an associated direction of travel of the vehicle for the data sample, and the association of the data sample with the road segment further includes the associated direction of travel of the vehicle for the data sample. 236. The method of claim 235, wherein is based at least in part on corresponding to one or more directions of travel associated with the road segment.   For each of at least some of the plurality of data samples, the direction of travel associated with the data sample is determined by using a reported location indicated by a number of data samples corresponding to the vehicle of the data sample. 237. The method of claim 236 further comprising estimating.   236. The association of claim 235, wherein the association of a data sample to a road segment is further based at least in part on one or more movement characteristics of the vehicle for the data sample other than the reported location. the method of.   The association of data samples to a road segment is further the one or more associated with the road segment by one or more distances determined based at least in part on the accuracy of the reported location. 236. The method of claim 235, based at least in part on expanding the position of the.   The one or more determined distances by which the one or more locations associated with a road segment are expanded for a data sample are based at least in part on the type of source of the data sample 240. The method of claim 239.   For each of one or more of the at least some road segments, the multiple related data samples of the road segment are each reported from a traffic sensor that monitors the road segment, each of the traffic 207. The method of claim 207, further comprising a number of data samples reflecting one or more locations on the road segment corresponding to a sensor.   Receiving one or more indications of a plurality of data samples each reported from one of a number of traffic sensors monitoring a number of road segments, and wherein the plurality of data samples are reported from Conditioning at least some of the plurality of data samples to take into account a traffic sensor; and for each of at least some of the plurality of data samples, each of at least one of the road segments Associating the data sample with the at least one road segment based at least in part on the one or more positions reflected by the data sample corresponding to one or more positions associated with 242. The method of claim 241, comprising.   For each of one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is determined by the road segment Based on identifying multiple data samples reported by a single vehicle traveling over and combining information from the identified multiple data samples, the identified multiple data samples are represented. 207. The method of claim 207, comprising: determining not to.   For each one or more of the at least one road segment, each of the plurality of associated data samples of the road segment is further when the vehicle of the data sample is at the reported location. Reflecting the reported time, the automatic analysis of the multiple related data samples of the road segment is further such that the actual vehicle movement on the road segment is movement during a predetermined period of time. 207. The method of claim 207, wherein the method corresponds to the predetermined period of time.   For each of a number of distinct time periods, a number of the one road segment that reflects the reported position of the vehicle at the reported time on one of the road segments during the time period. Further comprising receiving associated data samples, wherein the automatic analysis is performed for the one road segment for each of the certain time periods based on the data samples whose reported time is during the certain time period. 207. The method of claim 207.   For each one or more of the at least one road segment, the determination of the one or more data samples of the road segment that does not represent actual vehicle movement on the road segment is substantially 207. The method of claim 207, wherein the method is performed in real time.   At least some of the multiple data samples associated with at least some of the road segments are acquired by the vehicle prior to the reporting of the data samples by vehicles traveling on these roads; 247. The method of claim 246, wherein the reporting of at least some data samples occurs in a substantially real-time manner after acquisition of one or more of the at least one data sample. A computer readable medium, the contents of which enable a computing device to evaluate a data sample representing a moving vehicle by performing the method, the method comprising:
Receiving an indication of a number of data samples each reflecting a reported movement characteristic of one of a number of vehicles traveling on one or more roads, the reported movement of the data samples The characteristic reflects the receiving position of the vehicle; and
Automatically determining whether one or more of the plurality of data samples does not represent actual vehicle movement of interest on the one or more roads, Determining automatically based at least in part on the movement characteristics;
One or more representations of the data samples that are not determined to be represented in order to be usable for use in facilitating travel on the one or more roads. Providing a computer-readable medium.   The reported movement characteristics of each data sample include the reported location of the vehicle to which the data sample corresponds, and one or more data samples are of interest on the one or more roads The determination of whether the actual vehicle movement is not representative of the reported vehicle position of the one or more data samples does not correspond to an actual vehicle movement position on the one or more roads 249. The computer-readable medium of claim 248, comprising determining that the one or more data samples do not represent based at least in part.   The one or more data samples based at least in part on the reported vehicle position of the one or more data samples not corresponding to the position of actual vehicle movement on the one or more roads. The determination of not representing the determination determines that the reported vehicle position of the one or more data samples does not correspond to a predetermined position of interest of the one or more roads. 249. The computer readable medium of claim 249, comprising:   The reported movement characteristics of each data sample include a reported direction of travel of the vehicle to which the data sample corresponds, and one or more data samples have interest in the one or more roads The determination of whether the actual vehicle movement being represented does not represent an actual vehicle movement on the one or more roads where the reported vehicle movement direction of the one or more data samples is 256. The computer-readable medium of claim 248, comprising determining that the one or more data samples do not represent based at least in part on not corresponding to one or more directions of travel. Medium.   Based at least in part on the reported vehicle movement direction of the one or more data samples does not correspond to the one or more directions of actual vehicle movement on the one or more roads. The determination that the one or more data samples do not represent that the reported vehicle travel direction of the one or more data samples is of interest to the one or more roads. 262. The computer-readable medium of claim 251, comprising determining that it does not correspond to a predetermined direction of travel.   The determination of whether one or more data samples does not represent an actual vehicle movement of interest of the one or more roads is at least one data for each of the one or more vehicles. Identifying one or more activities involving these one or more vehicles based on the reported travel characteristics from a sample, and the identified activities are uninteresting activities 253. The computer-readable medium of claim 248, comprising: determining that the at least one data sample does not represent based at least in part on   249. The computer readable medium of claim 248, wherein the computer readable medium is a memory of a computing device.   249. The computer readable medium of claim 248, wherein the computer readable medium is a data transmission medium that transmits a generated data signal including the content.   249. The computer-readable medium of claim 248, wherein the content is instructions that when executed cause the computing device to perform the method. A computing system configured to evaluate a data sample representing a moving vehicle, comprising:
A first component configured to receive, for each of a number of roads, an indication of a number of data samples of the road, each reflecting a position of a vehicle near the road;
For each of at least some of the multiple roads,
Automatically determining one or more of the multiple data samples of the road whose reflected vehicle position does not correspond to a vehicle movement in which the road is interested;
One or more of the multiple data samples of the road other than the determined data sample so that the data sample shown can be used for use in facilitating travel on the road And a data sample filter component configured to provide a display.   For each of one or more of the at least one multiple roads, the determination of the data samples of the roads whose reflected vehicle positions do not correspond to vehicle movements that are interested in the roads are: 258. The computing system of claim 257, comprising determining that these reflected vehicle positions do not match a predefined position of the road.   The data sample filter component is configured to generate an interest on the road based at least in part on one or more predefined travel directions of the road for each of one or more of the at least one multiple roads. And further configured to automatically determine one or more data samples of the road that reflect one or more vehicle travel directions that do not correspond to vehicle movement. 258. The computing system of item 257.   258. The computing system of claim 257, wherein the first component and the data sample filterer component each include software instructions that are executed in a memory of the computing system.   The first component comprises, for each of a number of roads, means for receiving an indication of a number of data samples of the road, each reflecting a location near the road, and the data sample filter component is Automatically, for each of at least some of the roads, one or more of the multiple data samples of the road whose reflected vehicle positions do not correspond to vehicle movements of interest of the road One of the multiple data samples of the road other than the determined data sample to determine and make the indicated data sample available for use in facilitating travel on the road, or 258. The computer of claim 257, comprising means for providing a plurality of displays. Ring system. In a computer-implemented method for facilitating movement on the road by providing reliable data reading of road traffic sensors associated with the road in a manner that accurately reflects actual vehicle movement on the road There,
Receiving an indication of a number of road segments of one or more roads, each road segment having said road traffic sensor providing data on the speed of vehicle movement by one or more associated road traffic sensors; Having a receiving step;
In order to facilitate movement on one or more roads by providing reliable data on vehicle movement, for each of at least some of the road traffic sensors,
Receiving a number of data readings from the road traffic sensor, each including a reported speed of the one or more vehicles traveling by the road traffic sensor at an associated time within a recent period of time;
Determining a current data reading distribution of the road traffic sensor to reflect a reported vehicle movement speed during the recent period based on the received data reading;
Determining an average historical data reading distribution of the road traffic sensor to reflect an average vehicle movement speed during one or more previous fixed periods corresponding to the recent fixed period, the average historical data Determining a reading distribution based on a number of data readings received from the road traffic sensor during the one or more previous time periods;
Determining a statistical measure of the entropy of each of the current data read distribution and the average historical data read distribution and a statistical measure of the similarity between the current data read distribution and the average historical data read distribution Generating a comparison of the current data reading distribution and the average historical data reading distribution of the road traffic sensor based at least in part on determining a value;
Whether the generated comparison indicates a difference between the current data reading distribution of the traffic sensor and the average historical data reading distribution sufficient to reflect a likely malfunction of the road traffic sensor Determining whether the road traffic sensor probably provided a reliable data read for the recent period of time based on at least in part, and the road traffic sensor is reliable during the recent period of time. For at least a portion of the road segment associated with the road traffic sensor in a manner that is not based on the received data read during the recent period of time, if it is determined that it probably did not provide a data read. Estimate a reliable vehicle speed for a recent period and receive the received for the recent period. Providing the estimated vehicle speed for use as a replacement for the subsequent data reading and automatically providing reliable vehicle movement speed data for the recent period of time. .   For each one or more of the at least some road traffic sensors, based at least in part on whether it is determined that the road traffic sensor probably provided a reliable data read in the recent period of time. 262. The method of claim 262, further comprising: determining a sensor health status of the road traffic sensor; and providing an indication of the determined sensor health status of the road traffic sensor. Method.   For each of one or more of the at least some road traffic sensors, the estimate of the reliable vehicle speed for the recent period of time of at least a portion of the road segment associated with the road traffic sensor comprises: Reported vehicle travel speed of a second road segment associated with the road segment associated with a road traffic sensor, vehicle movement predicted to occur on the road segment associated with the road traffic sensor during the recent period of time 262. The method of claim 262, based on at least one of prediction information reflecting speed and a historical average vehicle travel speed of the road segment associated with the road traffic sensor.   For each of one or more of the at least some road traffic sensors, the determination of whether the road traffic sensor probably provided a reliable data read for the recent period of time further comprises the road traffic sensor. The determined statistical measure of the respective entropy of the current data reading distribution of the sensor and the average historical data reading distribution and between the current data reading distribution and the average historical data reading distribution of the road traffic sensor; The automated classification is performed by a neural network, at least in part based on a possibly reliable automated classification using the determined statistical measure of similarity 262. The method according to item 262.   For each of one or more of the at least some road traffic sensors, the determination of whether the road traffic sensor probably provided a reliable data read for the recent period of time further comprises the road traffic sensor. 268. The method of claim 265, based in part on an indication of operational status provided by a sensor and whether the road traffic sensor probably provided a reliable data read during a previous period of time.   For each one or more of the at least some road traffic sensors, the one or more previous time periods corresponding to the recent time period are a day of the week and the time associated with the recent time period 276. The method of claim 266, comprising a number of time periods selected to coincide with at least one of the times associated with a recent time period.   Each of the at least some road traffic sensors includes a loop sensor embedded in the road, a motion sensor installed close to the road, a radar distance measuring device installed close to the road, and a road proximity One of the installed radio frequency identifier devices, each of the at least some road traffic sensors being configured to measure the speed of a vehicle moving by the road traffic sensor 262. The method of claim 262, wherein:   For each one or more of the at least some road traffic sensors, at least some of the multiple data readings received from the road traffic sensors are each moved by the road traffic sensor during a period of time. 262. The method of claim 262, further comprising displaying a reported number of vehicles to play and / or an operational status of the road traffic sensor.   For each one or more of the at least some road traffic sensors, the determined statistical measure of similarity between the current data read distribution and the average historical data read distribution is the current 262. The method of claim 262, wherein the method is based on a Kullback-Leibler information amount calculation between a data read distribution of and a mean historical data read distribution.   The recent period of time is part of a day, and the automatic provision of the reliable vehicle movement speed data for each of one or more of the at least some road traffic sensors is performed throughout the day 276. The method of claim 262, wherein the method is performed multiple times a day to provide reliable vehicle travel speed data readings for a continuous period of time. A computer-implemented method for providing reliable data reading from road traffic sensors regarding traffic conditions on one or more roads, comprising:
Receiving, for each of one or more road traffic sensors each having an associated location on an associated road, information relating to multiple data reads taken by said road traffic sensor during a period of time, each data Reading has an associated time and reflects one or more measurements of traffic conditions at the associated time at the associated location of the associated road of the road traffic sensor; and
For each of the one or more road traffic sensors, movement on the one or more roads is facilitated by automatically removing road traffic sensor data readings that are likely to be unreliable. for,
Automatically determining whether the large number of data readings taken by the road traffic sensor during the period is likely to be unreliable, the determination being made of these multiple data readings Automatically determining, based at least in part on an automated comparison of information regarding at least some of the information with information regarding a number of other data reads previously taken by the road traffic sensor;
The associated location of the associated road of the road traffic sensor during the fixed period if it is not determined that the multiple data readings taken by the road traffic sensor during the fixed period are likely to be unreliable Providing a display that uses these multiple data loads in representing actual traffic conditions at
The associated roads of the associated road of the road traffic sensor during the period of time when it is determined that the multiple data readings taken by the road traffic sensor during the period of time are likely to be unreliable. Automatically providing an indication using other estimated data in place of these multiple data readings in representing the actual traffic situation at a location, wherein the other estimated data is Providing automatically based at least in part on other road traffic data related to reading these multiple data.   For each of at least one of the one or more road traffic sensors, at least some of the multiple data reads with the information regarding the multiple other data reads previously taken by the road traffic sensor. Determining a sensor health status of the road traffic sensor for the period of time and providing an indication of the determined sensor health status of the road traffic sensor based at least in part on the comparison of information regarding 278. The method of claim 272, further comprising:   After determining that the sensor health status of a road traffic sensor for a period of time is unhealthy, during one or more later periods, by the road traffic sensor during these later periods 278. The method of claim 273, wherein the automatic determination of whether a data read that is likely to be unreliable is further based at least in part on the determined unhealthy status of the period of time.   For each of at least one of the one or more road traffic sensors, the automatic determination of whether the large number of data reads taken by the road traffic sensor during the predetermined period is likely to be unreliable. Determining a current data reading distribution of the road traffic sensor to reflect traffic conditions during the period of time based on the at least some multiple data readings of the road traffic sensor; Determining an average historical data reading distribution to reflect an average traffic situation during one or more previous time periods based on the plurality of other data readings taken by 273. The method of claim 272.   For each of the at least one road traffic sensor, the comparison of the information about the at least some multiple data reads with the information about the multiple other data reads previously taken by the road traffic sensor is: 276. The method of claim 275, comprising comparing statistical measurements of information entropy of the current data reading distribution and the average historical data reading distribution.   For each of the at least one road traffic sensor, the comparison of the information about the at least some multiple data reads with the information about the multiple other data reads previously taken by the road traffic sensor is: 276. The method of claim 275, comprising determining a statistical measure of similarity between the current data reading distribution and the average historical data reading distribution.   For each of the at least one road traffic sensor, the determined statistical measure of similarity between the current data reading distribution and the average data reading distribution is based on a calculation of a Kullback-Leibler information amount. 277. The method of claim 277, wherein:   For each of at least some multiple data reads with each of at least one of the one or more road traffic sensors and the information regarding the multiple other data reads previously taken by the road traffic sensor. 278. The method of claim 272, wherein the comparison of the information further comprises classifying the information regarding the at least some multiple data reads.   279. The method of claim 279, for each of the at least one road traffic sensor, the classification is performed by at least one of a neural network, a decision tree, and a Bayesian classifier.   For each of at least one of the one or more road traffic sensors, the other estimated data used in place of the multiple data reads taken by the road traffic sensor during the period of time is 278. The method of claim 272, further based at least in part on a combination of at least some other road traffic sensor data readings related to these multiple data readings.   For one of the at least one road traffic sensor, the at least some other road traffic sensor data read is one or more nearby roads located on the associated road of the road traffic sensor. 290. The method of claim 281 comprising data reading taken by a traffic sensor.   The one road traffic sensor is one of a number of traffic sensors associated with one of a number of road segments of the road associated with the one road traffic sensor, the one or more 283. The method of claim 282, wherein nearby road traffic sensors are part of the one road segment.   The one road traffic sensor is one of a number of traffic sensors associated with one of a number of road segments of the road associated with the one road traffic sensor, the one or more 283. The method of claim 282, wherein nearby road traffic sensors are part of one or more other road segments proximate to the one road segment.   For one of the at least one road traffic sensor, the at least some other road traffic sensor data readings are data readings taken by the road traffic sensor during one or more previous time periods. 290. The method of claim 281, wherein the one or more previous time periods are selected to at least partially match a time category associated with the time periods.   For each of at least one of the one or more road traffic sensors, the at least some other road traffic sensor data read is associated with the road traffic sensor on the associated road during the period of time. 281. The method of claim 281 comprising data samples from a mobile data source that moves near a location to be.   For each of at least one of the one or more road traffic sensors, the other estimated data used in place of the multiple data reads taken by the road traffic sensor during the period of time is And, based at least in part on prediction information reflecting traffic conditions predicted to occur during the fixed period at the related position of the related road of the road traffic sensor, the prediction information is the fixed period 273. The method of claim 272, wherein the method is generated slightly before the fixed period based in part on a current traffic situation at a time of generating the prediction information.   For each of at least one of the one or more road traffic sensors, the other estimated data used in place of the multiple data reads taken by the road traffic sensor during the period of time is And, based at least in part on forecast information reflecting traffic conditions predicted to occur during the period of time at the relevant location of the relevant road of the road traffic sensor, the forecast information is the forecast information 273. The method of claim 272, wherein current traffic condition data at a time to generate is generated well before a period of time that is not used as part of generating the forecast information for the period of time.   For one of the road traffic sensors, failing to receive information regarding at least some missing data readings taken by the one road traffic sensor during a period of time; Automatically providing a display using other estimated data instead of the missing data read in representing actual traffic conditions at the relevant location of the relevant road of one road traffic sensor 278. The method of claim 272, further comprising:   Whether, for each of at least one of the one or more road traffic sensors, it is determined that the multiple data reads taken by the road traffic sensor during the period of time are likely to be unreliable 278. The method of claim 272, further comprising automatically determining an operational state of the road traffic sensor based at least in part and providing an indication of the operational state.   For each of at least one of the one or more road traffic sensors, the automatic determination of whether the large number of data reads taken by the road traffic sensor during the predetermined period is likely to be unreliable. And a day of the week associated with the period of time, a time associated with the period of time, an indication of the operating conditions provided by the road traffic sensor, and the road traffic sensor is probably trusted during one or more previous periods of time. 273. The method of claim 272, wherein the method is based on whether a possible data read was provided and many of the absence of data reads normally taken by the road traffic sensor.   For each of at least one of the one or more road traffic sensors, the multiple data readings of the road traffic sensors are each for a vehicle moving by the road traffic sensor at the associated time of the data readings. 273. The method of claim 272, comprising a reported rate.   For each of at least one of the one or more road traffic sensors, the multiple data readings of the road traffic sensor respectively report a reported quantity of vehicles moving by the road traffic sensor over a period of time and / or 273. The method of claim 272, further comprising displaying an operational status of the road traffic sensor.   For each of at least one of the one or more road traffic sensors, further comprising providing a reliable data reading of the road traffic sensor to one or more traffic data clients, wherein the reliable data reading is 278. The method of claim 272, comprising at least some of the multiple data reads and / or the other estimated data.   Each of at least some of the one or more road traffic sensors is installed proximate to the associated road of the road traffic sensor, a loop sensor embedded in the associated road of the road traffic sensor. Motion sensor, a radar distance measuring device installed in proximity to the relevant road of the road traffic sensor, and a radio frequency identifier device installed in proximity to the relevant road of the road traffic sensor Each of the at least some road traffic sensors is configured to measure traffic conditions at the relevant location of the relevant road of the road traffic sensor. 273. The method of claim 272, wherein:   The method is performed a number of times daily to provide reliable data reading of at least some of the one or more road traffic sensors of each of a number of portions of the day. 273. The method of claim 272. A computer-readable medium, the contents enabling a computing device to provide reliable data reading from road traffic sensors regarding traffic conditions on the road by performing the method, the method comprising:
Receiving a number of data readings generated by traffic sensors associated with the road, each of which reflects one or more measurements of traffic conditions on the road associated with an associated time;
The current reliability of the traffic sensor based at least in part on a comparison of information regarding at least some of the multiple data reads with information regarding a number of other data reads previously generated by the traffic sensor Automatically determining
The determination of the traffic sensor for use in facilitating travel on the road so that data readings generated by currently unreliable traffic sensors are no longer used to represent actual traffic conditions Providing a displayed current reliability indication.   The information related to the at least some multiple data reads includes a first data read distribution based on the at least some multiple data reads, and relates to the multiple other data reads previously generated by the traffic sensor. 297. The computer readable medium of claim 297, wherein the information includes a second data reading distribution based on the multiple other data readings previously generated by the traffic sensor.   The comparison of the information relating to the at least some multiple data reads with the information relating to the multiple other data reads previously generated by the traffic sensor may include the first data read distribution and the second data read. Determining a statistical measure of similarity between the distribution and determining a statistical measure of each entropy of the first data reading distribution and the second data reading distribution. 298. The computer readable medium of claim 298.   294. The computer of claim 297, wherein the determination of the current reliability of the traffic sensor is further based at least in part on classifying the information regarding the at least some multiple data reads. A readable medium.   The associated time of the multiple data reads is during a current period of time, and the determination of the current reliability of the traffic sensor relates to the current period of time, and further includes one or more previous times 297. The computer readable medium of claim 297, based at least in part on automatic determination of the reliability of each of the traffic sensors over a period of time.   The method is further for use in representing an actual traffic situation on the associated road at the associated time when the determined current reliability of the traffic sensor is determined to be reliable. Providing at least some of the multiple data readings and on the associated road at the associated time if the determined current reliability of the traffic sensor is determined to be unreliable. 297. The computer-readable medium of claim 297, comprising: providing other estimated data for use in representing actual traffic conditions.   302. The computer-readable medium of claim 297, wherein the computer-readable medium is at least one of a memory of a computing device and a data transmission medium that transmits a generated data signal that includes the content.   297. The computer-readable medium of claim 297, wherein the content is instructions that when executed cause the computing device to perform the method. A computing device configured to provide reliable data from traffic sensors regarding traffic conditions on an associated road,
Memory,
One or more after receiving information generated by a road-related traffic sensor reflecting one or more measurements of traffic conditions on the relevant road at a number of distinct times during a period of time At least a comparison of the generated information with other information previously generated by the traffic sensor to reflect one or more measurements of traffic conditions on the associated road over a period of time Based in part on a first module configured to automatically determine the reliability of the generated information in representing actual traffic conditions on the associated road during the period of time;
Actual traffic on the associated road during the period of time to facilitate travel on the associated road through the use of information that reliably represents actual traffic conditions on the associated road. And a second module configured to provide an indication of the determination of the reliability of the generated information in representing a situation.   The automatic determination of the reliability of the generated information in representing an actual traffic situation on the associated road during the period of time is such that the generated information is sufficient for the reliability of the period of time. Determining whether to reflect a minimum number of measurements to provide a degree, and the comparison of the generated information with other information includes determining that the generated information is the minimum number of measurements. 306. The computing device of claim 305, executed only when reflecting.   If the generated information does not reflect the minimum number of measurements to provide a sufficient degree of reliability for the period of time, the generated information is a portion of the road corresponding to the traffic sensor. Replaced with other estimated data based at least in part on other road traffic data related to the provided indication of the determination of the reliability of the generated information is the other estimated 302. The computing device of claim 305, comprising providing a display of the data that has been recorded.   The provided indication of the determination of the reliability of the generated information represents an actual vehicle movement on the road during the period of time if the generated information is determined to be reliable. Includes an indication for using the generated information, and if the generated information is not determined to be reliable, another estimated to represent actual vehicle movement on the road during the period of time. 306. The computing device of claim 305, comprising a display for using the data.   306. The computing device of claim 305, wherein the first module and the second module include software instructions that are executed in the memory.   The first module has received information generated by a road-related traffic sensor that reflects one or more measurements of traffic conditions on the associated road at a number of distinct times during a period of time. Later, the generated with other information previously generated by the traffic sensor to reflect one or more measurements of traffic conditions on the associated road for one or more other time periods Means for automatically determining the reliability of the generated information in representing actual traffic conditions on the associated road during the period of time, based at least in part on a comparison of the information, The second module comprises means for providing an indication of the determined reliability of the generated information in representing actual traffic conditions on the associated road during the period of time. The computing device of claim 305, wherein the door.

Images