CN102394008B - Assessing road traffic conditions using data from mobile data sources - Google Patents

Assessing road traffic conditions using data from mobile data sources Download PDF

Info

Publication number
CN102394008B
CN102394008B CN201110221617.2A CN201110221617A CN102394008B CN 102394008 B CN102394008 B CN 102394008B CN 201110221617 A CN201110221617 A CN 201110221617A CN 102394008 B CN102394008 B CN 102394008B
Authority
CN
China
Prior art keywords
segment
data sample
road
data
traffic
Prior art date
Application number
CN201110221617.2A
Other languages
Chinese (zh)
Other versions
CN102394008A (en
Inventor
克雷格·H·查普曼
亚历克·巴克
米切尔·A·小博恩斯
罗伯特·C·卡恩
奥利弗·B·唐斯
杰西·S·赫奇
斯科特·R·兰弗
Original Assignee
因瑞克斯有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US77894606P priority Critical
Priority to US60/778,946 priority
Priority to US78974106P priority
Priority to US60/789,741 priority
Priority to US11/431,980 priority
Priority to US11/432,603 priority
Priority to US11/432,603 priority patent/US20070208501A1/en
Priority to US11/431,980 priority patent/US20070208493A1/en
Priority to US11/438,822 priority patent/US7831380B2/en
Priority to US11/438,822 priority
Priority to US11/444,998 priority
Priority to US11/444,998 priority patent/US8014936B2/en
Priority to US11/473,861 priority
Priority to US11/473,861 priority patent/US7912627B2/en
Priority to US60/838,700 priority
Priority to US83870006P priority
Priority to US11/540,342 priority
Priority to US11/540,342 priority patent/US7706965B2/en
Application filed by 因瑞克斯有限公司 filed Critical 因瑞克斯有限公司
Priority to CN200780015916.22007.03.02 priority
Publication of CN102394008A publication Critical patent/CN102394008A/en
Application granted granted Critical
Publication of CN102394008B publication Critical patent/CN102394008B/en

Links

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions

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

Use the data estimation road traffic condition from Mobile data source

The divisional application of the present invention's to be application number be patented claim of 200780015916.2 (" using the data estimation road traffic condition from Mobile data source ").

Technical field

Following open text relates generally to a kind of data obtained from various data source to estimate the technology of road traffic condition, such as, by inferring the information about traffic on these roads based on the data sample of the actual travel reflected on road interested.

Background technology

Due to road traffic with than road capacity larger speed continue to increase, the traffic congestion of surge produces ill effect to business and government operation and individual happiness.Therefore, carried out various effort in every way to resist the traffic congestion of surge, such as by obtaining the information of current traffic condition and information being supplied to individuals and organizations.Can by various mode (such as, via radio-frequency (RF) broadcast, internet site, internet site shows the map of geographic area, on some main roads of wherein this geographic area, current traffic congestion is represented by color-coded information, and information can be sent to cellular mobile phone and other portable consumer device etc.) such current traffic condition information is supplied to interested parties.

A kind of source obtained about current traffic condition information comprises the observation that manually provides (such as, helibus about the magnitude of traffic flow and accident general information is provided, the report etc. of being sent via mobile phone by driver), and in the metropolitan area that some are larger, another kind of source is traffic sensor network, it can measure the magnitude of traffic flow (such as, by being embedded in the sensor in pavement of road) of different kinds of roads in the zone.Although the observation manually provided can provide some values in a limiting case, such information is usually only limitted to a few regions at every turn and usually lacks the enough details being enough to use.

In some cases, traffic sensor network can provide the more detailed information of some road traffic conditions.But there is various problem in such information and the information provided by other similar source.Such as, a lot of road does not have path sensor (such as, there is no the geographic area of path sensor and/or be not large enough to there is path sensor and as the arterial highway closing on a network part), the road even with path sensor also may often can not offer precise data, and this greatly weakens the data value that traffic sensor provides.A kind of reason of non-precision and/or non-authentic data comprises traffic sensor and damages, thus can not provide data, or provides intermittent data, or correctly can not read data.The another kind of reason of non-precision and/or non-authentic data is included in the problem that one or more sensor carries out data temporary transmission, causes being interrupted transmitting, or postpones to transmit, or does not transmit data.In addition, a lot of sensor do not configure or the person's state that designs to report surrounding driver (such as, whether their function is normal), even if the status information reporting driver also may incorrect (such as, report driver function normally but in fact really not so), so just very difficultly maybe can not determine that whether the data provided by traffic sensor accurate.In addition, about the information of traffic only can obtain with original and/or discrete form, thus use limited.

Hide, provide a kind of technology of improvement to obtain and estimate information about traffic and provide various relevant additional ability to be very helpful.

Accompanying drawing explanation

Fig. 1 be a diagram that the block scheme of the data stream between the assembly for estimating the embodiment of the system of road traffic condition at least in part based on the data obtained from vehicle and other Mobile data source.

Fig. 2 A-2E illustrates the example estimating road traffic condition at least in part based on the data obtained from vehicle and other Mobile data source.

Fig. 3 be a diagram that the block scheme of the computing system of data sample management system (the Data Sample Manager System) embodiment be suitable for described by execution.

Fig. 4 is the process flow diagram of the exemplary embodiment of data sample filterer device routine.

Fig. 5 is the process flow diagram of the exemplary embodiment of data sample exceptional value remover (Outlier Eliminator) routine.

Fig. 6 is the process flow diagram of the exemplary embodiment of data sample speed estimator routine.

Fig. 7 is the process flow diagram of the exemplary embodiment of data sample flow estimation device routine.

Fig. 8 is the process flow diagram that Mobile data source information provides the exemplary embodiment of routine.

Fig. 9 A-9C illustrates and obtains and the action example provided about Mobile data source in road traffic condition.

Figure 10 A-10B illustrates the example revising the data sample obtained from road traffic sensors.

Figure 11 is the process flow diagram of the exemplary embodiment of sensing data read error detector routine.

Figure 12 is the process flow diagram of the exemplary embodiment of sensing data read error corrector routine.

Figure 13 is the process flow diagram that sensing data reads the exemplary embodiment of gatherer routine.

Figure 14 is the process flow diagram of the exemplary embodiment of magnitude of traffic flow estimation device routine.

Embodiment

Based on the data that the traffic obtained is relevant, the technology of estimation road traffic condition is described in every way, the vehicle such as coming comfortable road travels is with other Mobile data source and/or from traffic sensor (such as, being embedded in physical sensors in road or neighbouring).In addition, at least some embodiments, the data sample come from Mobile data source with the data filling from other source one or more, can such as pass through the data of the physical sensors reading obtained in road annex or road.Based on obtained data sample (such as, from road traffic sensors, from each Mobile data source or collect the data that data point reads) various filtration and/or the adjustment of data sample and reading can be comprised to the estimation of road traffic condition, and the various deduction of interested traffic correlated characteristic and probability are determined.

As described in, road traffic condition information data obtained in certain embodiments can comprise by Mobile data source (such as, vehicle) multiple data samples of providing, from the data readings of the traffic sensor (being such as embedded in the loop sensor in pavement of road) based on road, with from the data of other data source.Data can be analyzed so that determine interested traffic feature in the various modes of the total vehicle total amount estimated in the specific part of the average traffic speed such as estimated and interested road etc., so that with in real time or be bordering on the determination that the mode (such as receiving bottom data sample and/or reading) in real time performs traffic.Such as, the data obtained can adjust to detect and/or correct mistake in the data in every way.If the road traffic condition information obtained coarsely maybe can not represent interested actual traffic condition characteristic, then can also carry out in every way filtering to remove data in various embodiments, comprise by using at least part of non-interested data sample based on road with the data sample associated according to other data sample and/or be considered as identical as the data sample adding up exceptional value, in certain embodiments, filter to comprise and perform associating data sample and specified link.Data sample after filtration can also comprise other and reflect vehicle location or non-interested behavior (such as, the vehicle berthed, vehicle spins in parking lot or building) data sample and/or other can not represent the data sample that on interested road actual vehicle travels.Estimate that the data obtained can comprise the traffic (such as, the magnitude of traffic flow and/or average traffic speed) determining road network various piece in specific geographical area at least in part based on obtained data sample at least some embodiments.Then can use estimated data to perform and relate to prediction, forecast, and/or other function of traffic-relevant information is provided.In at least some embodiments, data sample management system uses the technology described by least some to prepare the data used by traffic data clients, the predicted traffic information such as producing multiple forecasts of traffic in multiple time in future provides system, and this will in following detailed description.

In certain embodiments, the adjustment of fetched data sample can comprise the data sample corrected mistakes, such as by detecting and/or correct the mistake (such as, from the data readings that road traffic sensors receives) in current data in every way.Particularly, such as based on the analysis of the data sample provided by these data sources, the technology of " health " for estimating particular source (such as based on the traffic sensor of road) is described to determine whether data source is working properly and reliably to provide precise information sample.Such as, in certain embodiments, the former data readings current data reading provided by given traffic sensor and this traffic sensor provided (such as, historical average certificate) compare, to determine that whether Current traffic data readings is significantly different from former common data readings, such as this can be caused by the other problem in this traffic sensor non-normal working and/or data, and/or can replace to reflect abnormal current traffic condition.This determination and analysis to possible errors in particular source and/or Current traffic data readings can be performed in various embodiments in every way, this more will discuss in detail following, comprise at least partly based on the sorting technique such as using neural network, Bayes classifier, decision tree etc.

After detecting the corrupt data sample such as arrived from the damaged data source of abnormal work, can correct or revise this corrupt data sample (and the data sample lost) by this way.Such as in certain embodiments, one or more data source can be revised (such as by of use relevant information or other, traffic sensor) obliterated data and corrupt data, such as by from close on or other normal work relevant traffic sensor while data sample (such as, by taking the mean to the data readings provided by adjacent traffic sensor), by relating to the foresight information of loss and corrupt data sample (such as, expected data reading by using the foresight of these data sources and/or predictive traffic related information to determine one or more data source), via one or more data source historical information (such as, by using historical average according to reading), adjust via the relevant consistent deviation of use or other type of error that can compensate that leads to errors with correction data sample etc.Relating to other details revising loss and corrupt data sample will in following detailed description.

In addition, the technology of description, also for various alternate manner estimation traffic related information, such as allows the situation of the correction of the data sample reliably performing particular source (such as, special traffic sensor) in current available data.Such as, the existence of the unhealthy traffic sensor of multiple abnormal work may cause not having enough data to estimate traffic flow information fully credibly to each in these traffic sensors.In this case, traffic related information may be estimated with various alternate manner, comprises based on relevant traffic sensor group and/or the out of Memory relating to road network structure.Such as; to be described in more detail as following; each interested road can come modeling or expression by using multiple road segment segment, the data that each road segment segment can have the traffic sensor of multiple association and/or obtain from other data source one or more (such as, Mobile data source).The words of if so, road traffic condition information can be estimated in every way for particular lane section (or other group of multiple relevant traffic sensor), such as by using the traffic related information for estimating neighboring track section, for particular lane section information of forecasting (such as, produced in limited future time section at such as three hours etc., at least in part based on situation recent in current and the schedule time), to the forecast information in particular lane section (such as, produce in such as two weeks or future time section for more time, not use some or all of the current and recent condition information for predicting), the history long-run average etc. in particular lane section.By using such technology, even if also can traffic related information be provided when only having the current traffic condition data of a small amount of or neither one or multiple approaching sensor or other data source.Other details relating to such traffic related information estimation will in following detailed description.

As previously mentioned, the information of electronic map traffic can obtain in every way from Mobile data source in various embodiments.In at least some embodiments, Mobile data source comprises the vehicle on road, and it is each comprises one or more computing system providing associated vehicle Mobile data.Such as, every portion vehicle can comprise GPS (" GPS ") equipment and/or other geolocation device can determined geographic position, speed, direction and/or other sign or relate to the data that vehicle travels, and the one or more equipment (no matter whether being geolocation device or disparate communications apparatus) sometimes on vehicle can by such data (such as, pass through wireless link) be supplied to the system of the such data of one or more energy use (such as, data sample management system, will in following more detailed description).Such as, such vehicle can comprise the distributed network by the vehicle of each incoherent user operation, fleet (such as, for express company (delivery company), taxi and bus company, carrier, government department or agency, the vehicle etc. of car rental services), be subordinate to and provide relevant information (such as, OnStar serves) the vehicle of commercial network, be operated to the vehicle group of the such traffic related information of acquisition (such as, by travelling predetermined route, or traveling dynamically changes direction on road, to obtain the information about interested road), be mounted with the vehicle of mobile telephone equipment (such as, as built-in device and/or have vehicle-mounted thing (vehicle occupant)) positional information can be provided (such as, based on the GPS ability of equipment and/or based on the geo-positioning capabilities provided by mobile network) etc.

In at least some embodiments, Mobile data source can comprise or based on other mobile device of the user that computing equipment and road travel, such as user is driver and/or the passenger of vehicle on road.Such subscriber equipment can comprise the equipment (such as, mobile phone and other handheld device) with GPS ability, or position and/or mobile message alternatively also can otherwise produce in other embodiments.Such as, equipment in vehicle and/or subscriber equipment can carry out communicating (such as with the external system of energy detection and tracking relevant device information, by the multiple emittor/receivers in the network of Dynamic System each via equipment), thus the position of equipment and/or mobile message are determined in the various modes with various level of detail, or such external system can also detection and tracking associated vehicle and/or user information and not with equipment mutual (such as, can observe and identify the camera system driving board and/or user's face).Such as, such external system can comprise mobile phone tower and network, other wireless network (such as, Wi-Fi Hotspot), uses the detecting device of the vehicle transducer of the various communication technology (such as, RFID, or " radio frequency identification "), other detecting device of vehicle and/or user (such as, uses infrared ray, sonar, radar or Laser Distance Measuring Equipment are to determine position and/or the speed of vehicle) etc.

The road traffic condition information obtained from Mobile data source can be used in every way, no matter separately or use together with other road traffic condition information from one or more other originate (such as, from road traffic sensors).In certain embodiments, use such road traffic condition information obtained from Mobile data source, info class is provided to be similar to data from path sensor, but for there is no the road of the path sensor run (such as, for the road lacking sensor, such as there is no the geographic area of networks of road sensors and/or the arterial highway not even as big as there being sensor, path sensor etc. for damaging), with the Copy Info that verification receives from path sensor or other source, thus identification provides the path sensor of non-precision data (such as, due to interim or current problem) etc.And road traffic condition can be measured in one or more ways or represent, no matter be based on the data sample from Mobile data source and/or traffic sensor data readings, such as, in absolute (such as, average velocity; The volume of traffic in the indicated time period; The average holding time of other position on one or more traffic sensor or road, such as with represent vehicle by or the average percentage of activated sensors time; The calculating grade of one or more congestion in road, such as, measure based on other traffic one or more; Etc.) and/or (such as, represent the difference with normal conditions or maximum case) in relative.

In certain embodiments, some road traffic condition information can take the form of the data sample provided by various data source, and the data source such as associated with vehicle is with the travelling characteristic of reporting vehicle.Each data sample can comprise the quantity of information of change.Such as, the data sample provided by Mobile data source can comprise one or more source identifier, speed identifier, orientation or direction, position instruction, timestamp and status identifier.Source identifier can be mark as the numeral of the vehicle (or people and miscellaneous equipment) of data source or string.In certain embodiments, Mobile data source identifier can with Mobile data source permanent or temporary association (such as, for the life-span in Mobile data source; For one hour; For the session of current use, such as, so that unlocking vehicle or data-source device just distribute a new identifier each time).In at least some embodiments, source identifier associates with Mobile data source, to make the secret relation related to from the data in Mobile data source minimize (no matter being permanent or temporary association), such as, by identify based on identifier to create and/or operate source identifier with the mode in the Mobile data source of this Mobile data source and identifier linkage to stop.Speed instruction can reflect the instant of the Mobile data source represented in every way or average velocity (such as, mph.).Orientation can reflect the direction of traveling, and is the angle or other tolerance (such as, based on orientation or the radian of compass) that represent with " degree ".Position instruction can reflect the physical location (such as lat/lon to or Universal Transverse Mercator coordinate) represented in every way.Timestamp can indicate the time of Mobile data source record sample preset time, such as, with local zone time or UTC (" Universal Coordinated Time ") time.Status identifier can represent the state in Mobile data source (such as, vehicle moving, stop, engine running and stops) and/or at least some state (such as, electricity is low, signal intensity is weak) of sensing, record and/or transmitter.

In certain embodiments, the road network in given geographic area can come modeling or expression by using multiple road segment segment.Each road segment segment may be used for the part representing road (or multiple road), such as by given physical road is divided into multiple road segment segment (such as, each road segment segment has specific length, the road of such as one mile long, or select to reflect that the road sections of similar traffic feature is as road segment segment), multiple road segment segment like this can be road continuous print parts, or alternatively in certain embodiments, they can the overlapping or all mutual part disturbed of any road segment segment.In addition, road segment segment can represent the one or more traveling lanes on given physical road.Therefore, both direction each on have the specific multilane of one or more traveling lane can associate with at least two road section, wherein at least one road segment segment associates with the traveling on a direction, and at least another associates with the traveling on other direction.In addition, in some cases, multiple tracks of the single road travelled in a single direction can be represented by multiple roadway segment, if such as track has different travel conditions features.Such as, given freeway facility can have quick or high occupancy (" HOV ") track, it can be represented using as quick or HOV track by the mode far different with representing routine (such as, non-HOV) track that equidirectional travels.Roadway segment can also be connected to other adjacent road segment segment or the road segment segment adjacent with other associates, thus forms road segment segment network.

Fig. 1 be a diagram that at least in part based on the embodiment of the system of the data estimation road traffic condition obtained from vehicle and other Mobile data source assembly between the process flow diagram of data stream.Shown data flowchart is intended to be reflected in data source, i.e. the assembly of the embodiment of data sample management system, and the logical expressions of data stream between traffic data clients.That is, actual data stream may occur via various mechanism, comprise the indirect stream of direct stream (such as, by being realized by parameter or the network service of such as message) and/or the storage system via one or more Database Systems or other such as file system.Shown data sample management system 100 comprises data sample exceptional value and removes assembly 106, data sample speed assessor component 107, data sample stream estimation assembly 108 and Alternative sensors collection assembly 110.

In an illustrated embodiment, the assembly 104-108 and 110 of data sample management system 100 obtains data sample from various data source, and this comprises based on the data source 101 of vehicle, road traffic sensors 103 and other data source 102.Data source 101 based on vehicle can be included in multiple vehicles that one or more road travels, its each miscellaneous equipment that can comprise one or more computing system and/or can provide associated vehicle running data.As being described in more detail in addition, every portion vehicle can comprise GPS and/or can determine the geolocation device of position that associated vehicle travels, speed and/or other data.Such data can by the assembly of described data sample management system by wireless data link (such as, satellite uplink and/or mobile telephone network) or alternate manner is (such as, after vehicle arrives certain physical location, such as carry out after its base is got back to by fleet physics wired/cable connect) obtain.Road traffic sensors 102 can comprise and is arranged on multiple sensors in each street, highway or other road, upper or neighbouring, and the loop sensor be such as embedded in road surface can measure time per unit by the vehicle fleet size on this sensor, car speed and/or other data relating to the magnitude of traffic flow.Data can obtain from road traffic sensors 102 via based on wired or wireless data link similarly.Other data source 103 can comprise the data source of other type various, the Map Services and/or database that there is provided about road network information are provided, link such as between road and relate to the traffic control information (such as, the existence of traffic control signal and/or position and/or speed limit region) of this road.

Although data sample is directly supplied to each assembly 104-108 and 110 of data sample management system 100 by the data source 101-103 in this example, in other embodiments data sample also can before being provided to these assemblies advanced row relax.Such process can comprise based on time, position, geographic area and/or individual data source identity (such as, vehicle, traffic sensor etc.) tissue and/or collect data sample in logical collection.In addition, such process can comprise merging or data splitting sample to more senior logical data sample or other value.Such as, the data sample obtained from the road traffic sensors of multiple geographically colocated can integrate with single logical data sample by average or other collection mode.In addition, such process can comprise to derive based on one or more obtained data sample or the element of generated data sample or data sample.Such as, in certain embodiments, at least some can provide based on each of data source of vehicle the data sample only comprising source identifier and geographic position, if so, so using specified time interval or section and multiple differing data sample groups of periodically providing just can associate with another and provide as particular vehicle At All Other Times.Can also process further such data sample group determine other about travel information, such as the orientation of each data sample (such as, by calculating in the position of data sample and the angle previously and/or between the position of subsequent data sample) and/or each data sample speed (such as, by calculating in the position of data sample and the distance previously and/or between the position of subsequent data sample, and by distance divided by the corresponding time).

In an illustrated embodiment, data sample filterer component 104 obtains data sample from the data source 101 based on vehicle with other data source 102, and they is being supplied to data sample exceptional value removal assembly 106 and is being supplied to before data sample stream estimates assembly 108 obtained data sample filterer alternatively.As will be more discussed in detail elsewhere, such filtration can comprise: associated with the road segment segment corresponding to road in geographic area by data sample, and/or identifies not corresponding to interested road segment segment or the data sample reflecting uninterested vehicle location or behavior.Data sample is associated with road segment segment and can comprise: use the reported position of each data sample and/or orientation to determine that whether this position and orientation are corresponding to previously defined road segment segment.Identify and can not comprise corresponding to the data sample of interested road segment segment: remove or identify such data sample so as not to their modelings, to consider or by other assembly process of data sample management system 100, the such data sample removed can comprise those roads corresponding to the road class of uninterested specific function (such as, residential street) data sample, those parts corresponding to uninterested road or region data sample (such as, ramp and collector/distribution lane/tell highway road) etc.Whether identification data sample reflects that uninterested vehicle location or behavior can comprise: identify and be in idle condition (such as, engine opens parking), drive corresponding data sample such as the vehicle of (such as, spinning with low-down speed) etc. in garage parking.In addition, in certain embodiments, filtration can comprise for presenting or analyze further and identifying that road segment segment is that (or not being) is interested.Such as, such filtration can comprise to be analyzed at special time period (such as, hour, day, week) changeability of the interior magnitude of traffic flow and/or the degree of blocking up of each bar road segment segment, there are (intra-time period) changeability in the low time period and/or low some or all road segment segment (such as, or their functional category of roads unavailable for sensor data readings represents the road segment segment of less or less travel) of blocking up as uninterested road and road segment segment to get rid of from analyze further.

The auxiliary data sample corrected mistakes of sensing data adjuster 105, such as, by detecting and correct the mistake of the reading obtained from road traffic sensors 103.In certain embodiments, adjusting component detection by sensing data is that insecure data sample is not forwarded to other assembly to use and (or provides the non-reliable expression of particular data sample, so that other assembly can process these data samples), such as, data sample exceptional value remover 106 is not forwarded to.If so, data sample exceptional value removes assembly then can determine whether that enough authentic data samples can be used, if not, then initiate correction behavior.Alternatively, in some embodiments and environment, sensing data adjustment assembly can also perform some to data sample and correct, ground will be more discussed in detail as following, then the data after correction are supplied to sensor collection assembly 110 (and being supplied to other assembly alternatively, such as data sample exceptional value removal assembly and/or data sample stream estimation assembly).Detect misdata sample and can use various technology, comprise statistical measurement, the historical rethinking of the distribution of the current data sample reported by given road traffic sensors to the data sample reported by this road traffic sensors within corresponding time period (such as, identical week number of days with one day in identical time) is compared.Difference that is actual and historical rethinking scope can be calculated by statistical measures, and such as Kullback-Leibler divergence, which provides the convex measuring of the similarity between two probability distribution, and/or statistical information entropy.In addition, some path sensors can the instruction of report sensor health, can also use such instruction to detect the mistake of obtained data sample.If detect mistake in obtained data sample, then can revise the data sample of makeing mistakes in every way, comprise utilize from determine error-free adjacent/mean value of the data sample on adjacent (such as, the side) of side path sensor replaces such data sample.In addition, can replace by using the value previously or simultaneously predicted and/or predict such as provided by predictive traffic information systems, revising the data sample of makeing mistakes.Other details that relating to predicted traffic information provides will provide in addition.

Data sample exceptional value is removed assembly 106 and is obtained the data sample after filtration from data sample filterer component 104 and/or obtain adjustment or revised data sample from sensing data adjustment assembly 105, then identifies and considers to remove the data sample that those do not represent the actual vehicle traveling in interested road and road segment segment.In an illustrated embodiment, for each interested road segment segment, block analysis in special time period record and the data sample group (such as, by data sample filterer component 104) associated with road segment segment, to determine if will remove, which should be removed.Can perform such determination to non-representative data sample in every way, comprise based on following technology: relative to other data sample in data sample group, detecting data sample is statistics exceptional value.Other details relating to the removal of data sample exceptional value will provide in addition.

Data sample speed assessor component 107 is removed assembly 106 from data sample exceptional value and is obtained data sample, so that the data sample obtained in the embodiment shown represents that the actual vehicle on interested road and road segment segment travels.Data sample speed assessor component 107 then analyzes the data obtained, with based on this road segment segment (such as, by data sample filterer component 104, or the reading by coming from the sensor of road segment segment part) the data sample group that associates with the time period, one or more speed of the road segment segment interested to estimating in the time period interested at least one.In certain embodiments, the speed estimated can comprise the speed average of the multiple data sample of this group, also can (such as, the age (age), so that the weighting of giving that newer data sample is larger by one or more attribute weights of data sample; And/or the source of data sample or type, so that from Mobile data source or the weighting that changes data sample from path sensor to the larger weighting in the source with higher expected reliability or availability).The more details relating to the velocity estimation carried out from data sample will provide elsewhere.

The data sample stream estimation time period of assembly 108 interested at least one is interested road segment segment estimation telecommunication flow information, to estimate the volume of traffic (such as, be expressed as to arrive in such as per minute or special time amount hourly or through the vehicle total amount of road segment segment or average), estimation traffic density (such as, be expressed as such as every mile or kilometer etc. the vehicle average of per unit distance or total amount) and estimation occupation due to communication rate (such as, be expressed as the average or total time quantum taking specified point or region at special time amount vehicle such as per minute or per hour etc.) etc.In an illustrated embodiment, to the estimation of telecommunication flow information at least in part based on the information relating to traffic speed removed assembly 106 by data sample speed assessor component 107 and data sample exceptional value and provide, alternatively can based on the traffic data sample information being adjusted assembly 105 and data sample filterer component 104 by sensing data and provide.Other details relating to the estimation of traffic sample stream will provide elsewhere.

If existed, then such as after sensing data adjustment assembly has eliminated any insecure data sample and/or have modified any loss and/or non-authentic data sample, the sensor-based traffic related information provided by sensing data adjustment assembly 105 collected by sensor data collection assembly 110.Alternatively, in other embodiments, sensor data collection assembly alternatively can perform removal and/or the correction of this loss and/or corrupt data sample.In some cases, the information that sensor data collection assembly 110 can be provided by the multiple independent traffic sensor associated with each road segment segment by collecting (such as, average) is these road segment segment eachly provides telecommunication flow information.Similarly, if existed, sensor data collection assembly 110 can provide information, estimate by such as data sample speed assessor component 107 and/or data sample stream the estimation traffic that the assembly of assembly 108 etc. provides to supplement, or can at the data sample from Mobile data source reliable or do not have the authentic data sample of q.s to allow such as data sample speed assessor component 107 and data sample stream to estimate other assembly of assembly 108 etc. provides when accurate estimation road traffic condition information alternatively uses.

One or more traffic data clients 109 obtains and estimates the road traffic condition information of the estimation that assembly 108 provides (such as by data sample speed assessor component 107 and/or data sample stream in an illustrated embodiment, speed and/or flow data), and such data can be used in every way.Such as, the traffic information system that traffic data clients 109 can comprise other assembly and/or be operated by the operator of data sample management system 100, such as foresight traffic information providing system, uses traffic related information to produce the traffic related information forecast at the future traffic condition of multiple future time; And/or in real time the transport information of (or being bordering in real time) presents system and obtains or provide system, provides in real time the traffic related information of (or being bordering on real-time) to terminal user and/or third party's client.In addition, traffic data clients 109 can comprise the computing system that operated by third party to provide transport information to its client.In addition, in some environment (such as, working as and can not obtain enough data to perform accurate estimation for data sample speed assessor component and/or data sample stream estimation assembly, and/or when from can not obtain data based on vehicle or other data source) this one or more traffic data clients 109 obtains the road traffic condition information provided by sensor data collection assembly 110 alternatively, the data from data sample speed assessor component and/or data sample stream estimation assembly can be substituted, or additionally obtain outside this.

In order to illustrated object, some embodiments, in following description, wherein estimate the road traffic condition of particular type in a particular manner, and use such estimation transport information in various specific mode.But, should be understood that, can otherwise also use the input data of other type in other embodiments to produce such traffic estimation, described technology can use in other situation widely, and the present invention is not limited to provided exemplary details.

Fig. 2 A-2E illustrates the example based on the data estimation road traffic condition obtained from vehicle and other Mobile data source, as performed by described data sample management system.Particularly, Fig. 2 A illustrates the example of data sample filterer, for having several roads 201,202,203 and 204 and having the example region 200 that instruction legend in a northerly direction indicates 309.In this example, road 202, the limited of such as highway or the highway that crosses enters road (limited access road), is divided in west to the different track group 202a and 202b with difference driving vehicle on east orientation.Track group 202a comprises HOV track 202a2 and multiple other conventional track 202a1, track group 202b comprise HOV track 202b2 and other conventional track 202b1 multiple similarly.Road 201 walks road 202 (such as, via overline bridge or bridge), and road 204 is onramps, and the northern runway 201b of road 201 is connected to the eastbound carriageway group 202b of road 202 by it.Road 203 is local frontage roads of adjacent road 202.

Road shown in can representing in every way in fig. 2, for described data sample management system.Such as, one or more road segment segment can associate with each physical road, such as, associated with southern runway 202b with northern runway 201a respectively with section, southern trade by north row.Similarly, at least one road segment segment that heads west can associate with eastbound carriageway group 202b with the track group 202a that heads west of road 202 respectively with at least one eastbound road segment segment.Such as, the part in the eastbound carriageway group 202b east of road 201 can be the part phase independently road segment segment with the group 202b west, track that heads west of road 201, such as change (such as based on general road traffic condition or through being everlasting between road segment segment, because usual vehicle significantly flow into the track group 202b of road 201 from onramp 204, so in general cause larger blocking up on the track group 202b to road 201 east orientation).In addition, can one or more tracks group be decomposed in multiple road segment segment, if such as different tracks generally or often has different road traffic condition features (such as, give certain portions as the first section corresponding to track 202b1 track group 202b based on these tracks of enjoying similar traffic feature, and will there is different traffic feature thus as the second lane section corresponding to HOV track 202b2 due to it)-in other this situation, single road segment segment is only had to may be used for such track group, but estimate this track group road traffic condition time some data samples (such as, corresponding to those of the 202b2 in HOV track) can get rid of from use (such as removing assembly by data sample filterer component and/or data sample exceptional value).Alternatively, multiple tracks of multiple given road can be expressed as single road section by some embodiments, even if this track is in the opposite direction up train, such as, when road traffic condition is similar usually in the two directions---such as, frontage road 205a can have two contrary driving lanes, but can be represented by a road segment segment.Road segment segment can otherwise be determined at least in part at least some embodiments, such as, associate (such as, physical dimension and/or orientation and/or traffic-relevant information (such as, speed limit) with geography information.

Fig. 2 A also describe specified time interval or At All Other Times section (such as, 1 minute, 5 minutes, 10 minutes, 15 minutes etc.) period travels multiple data sample 205a-k that multiple Mobile data sources in region 200 (such as, vehicle, not shown) are reported.When a report by multiple Mobile data source, each of data sample 205a-k is illustrated as arrow, and it represents the orientation of data sample.Data sample 205a-k to be superimposed upon on region 200 to reflect that position that each data sample reports (such as by this way, represent with dimension and precision unit, such as based on GPS reading), its record data sample time can different from the physical location of vehicle (such as, due to the reading of out of true or mistake, or due to the intrinsic variable precision of used position sensing mechanism).Such as, data sample 205g shows the slightly northern position of road 202b, it can reflect the vehicle (such as, due to mechanical fault) beinging drawn to 202b2 north side, track, or it can be reflected in track 202b2 or other track eastbound direction on the non-precision position of vehicle of actual travel.In addition, single Mobile data source can be the source of data sample more more than shown data sample, if such as sample 205i and sample 205h are reported (such as by the single portion vehicle travelled along road 202 east orientation within the time period, by comprising the single transmission of the multiple data samples for multiple previous time point, so that every 5 minutes or every 15 minutes report data samples).About storing and providing the more details of multiple fetched data sample to be included in following content.

Data sample management system described in certain embodiments can filter obtained data sample, data sample be mapped to predetermined road segment segment and/or identify not corresponding to the data sample of interested road segment segment.In certain embodiments, if the preset distance of reported position in the road corresponding with road segment segment and/or track (such as, 5 meters) in, and its orientation is in the predetermined angular (such as plus or minus 15 degree) in the orientation in the road corresponding with this road segment segment and/or track, then data sample associates with road segment segment.Although can perform before data sample can be used for data sample management system the association of the data sample of road segment segment in other embodiments, road segment segment in illustrated embodiment and enough location-based information are (such as, the orientation of road segment segment, the physical extent etc. of road segment segment) association, to make such determination.

As directed example, data sample 205a can associate with the road segment segment corresponding to road 203 because its reported position drop on road 203 scope in and its orientation identical with at least one orientation of associated road 203 (or being bordering on identical).In certain embodiments, when using single road section to represent the multiple track travelled in the opposite direction, whether the orientation of data sample comparatively can be associated with this road segment segment with determining data sample with two aspect ratios of road segment segment.Such as, data sample 205k has roughly contrary with data sample 205a orientation, if but use road segment segment to represent two opposite carriageway of road 203, then it also can associate with the road segment segment corresponding to road 203.

But, because road 203 is close with track group 202a, also possibly, because the orientation of data sample 205k is identical with the orientation of track group 202a, then data sample 205k is reflected in the vehicle that track group 202a travels, if the blank space of vehicle location mistake that the reported position of such as data sample 205k travels in one or more tracks of track group 202a.In certain embodiments, multiple possible road segment segment can remove based on the out of Memory associated with this data sample for the situation of a data sample.Such as, in this case, the analysis of the reported speeds of data sample 205k can contribute to this removal, if such as track group 202a is corresponding to the highway of 65mph speed limit, road 203 is for having the local frontage road of 30mph speed limit, and the speed that data sample is reported is 75mph (causing large with the possibility associated than associating with local frontage road in highway track).In general, if the reported speeds of data sample 205k compares observation or transmission speed that the observation of track group 202a or transmission speed are more similar to road 203, then such information may be used for partly determining data sample to associate instead of track group 202a with road 203.Alternatively, if the reported speeds of data sample 205k is more similar to observation or the transmission speed of track group 202a compared to the speed of the road 203 observed or send, then it just associates instead of road 203 with track group 202a.Also a part (such as, the position in this removal is can be used as like the info class of other type; Orientation; State; Other relates to the information of data sample, other most recent data sample etc. such as come from identical Mobile data source), the matching degree of data sample information type and candidate roads section is such as reflected as the part of weighted analysis.

Such as, associate with the road segment segment be applicable to for by data sample 205b, the position that it is reported appears at the part of track 201b and track group 202a overlap, and it closes on track 201a and other road.But, the orientation (roughly north row) reported of data sample compared to the orientation of other candidate lane/road and the orientation (north is capable) of track 201b closer to, therefore it probably associates with the road segment segment corresponding to track 201b in this example.Similarly, data sample 205c comprises can mate multiple road/track (such as track 201a, 201b, with track group 202a) reported position, but its orientation (roughly heading west) may be used for selecting road segment segment for track group 202a as the most suitable road segment segment of this data sample.

Or this example, data sample 205d can not associate with any road segment segment because its orientation (roughly eastbound) with corresponding to the reported position of this data sample track group 202a (heading west) be in reverse direction.If the candidate roads section not having other suitable, the position that itself and data sample 205d report is enough near (such as, in predetermined distance), if the track group 202b such as with similar orientation is too far away, then during the follow-up use of the analysis from this data sample is filtered, get rid of this data sample.

Data sample 205e can with such as associating corresponding to the road segment segment of the road segment segment of HOV track 202a2 corresponding to track group 202a, this is because its reported position and orientation are corresponding to the position in this track and orientation, if the location-based technology such as the position of this data sample has enough resolution to distinguish track (such as, different GPS, infrared ray, sonar or radar ranging equipment).Data sample can also associate with the specific track of multiple-lane road, if such as track has different traffic features based on the factor except position-based information.Such as; in certain embodiments; can usage data sample reported speeds by the expection distribution of the speed observed the data sample for each such candidate lane (or other of the magnitude of traffic flow is measured) (such as; usually or Gaussian distribution) modeling, and data sample is conformed to specific track or mates.Such as, the speed reported due to this data sample compared to the observation of the vehicle travelled on conventional track 202a1, deduction or historical average speeds closer to the observation of the vehicle that HOV track 202a2 travels, deduction or historical average speeds, therefore data sample 205e can associate with the road segment segment corresponding to HOV track 202a2, such as by determining the analysis of observation or deduction speed (such as, using the data readings provided by one or more road traffic sensors) and/or other relevant current data based on other data sample.

In a similar fashion, data sample 205f, 205h, 205i and 205j can respectively with corresponding to track 201a, track 202a1, track 202b1 associates with the road segment segment on slope 204, because their positions of reporting and orientation are corresponding to the position in these roads or track and orientation.

Even if its reported position is outside the scope of shown road, data sample 205g also can associate (such as with the road segment segment corresponding to track group 202b, road segment segment for HOV track 202b2), this is because reported position can in the predeterminable range of road (such as 5 meters).Alternatively, if the reported position of data sample 205b is away from road, then it also can not associate with any road segment segment.In certain embodiments, use different predeterminable ranges can to the data sample provided by different pieces of information source, to reflect the accuracy level that data source is known or expect.Such as, by the data sample using the Mobile data source not correcting gps signal to provide can use relatively high (such as, 30 meters) predeterminable range, and the data sample that the Mobile data source correcting GPS device by usage variance provides comparatively speaking can use the predeterminable range of low (such as, 1 meter).

In addition, data sample filterer can comprise not corresponding with the interested road segment segment data sample of identification and/or can not represent the data sample of the actual vehicle travelled on road.Such as, some data samples can be removed, because they associate with the irrespective road of data sample management system according to considering.Such as, in certain embodiments, the data sample associated with the road (such as, residential block street and/or arterial highway) of secondary function road class can be filtered.Return Fig. 2 A again, such as, can filtering data sample 205a and/or 205k, because road 203 be positioned at low-down Function Classification local frontage road and not consider by data sample management system, or also can filtering data sample 205j, be not separated with expressway because onramp is too short.Filtration can also based on other factors, the such as deduction in other Mobile data source or report behavior in one or more road segment segment.Such as, associate with road segment segment and provided by single Mobile data source all represent that a series of data samples of same position likely represent that this Mobile data source stopped.If other data samples all associated with same link section all represent the Mobile data source of movement, data sample then corresponding to the Mobile data source stopped can owing to can not represent the actual vehicle that travels in this road segment segment by filtering, such as, because Mobile data source is the vehicle berthed.And, in certain embodiments, data sample can comprise the report instruction of vehicle driving state (such as, vehicle transfer is engine start " parking ", vehicle stops sending), if so, such instruction can be used similarly to carry out the such data sample that can not represent actual travel vehicle of filtering.

Fig. 2 B illustrate with at specified time interval or the view obtaining the multiple data samples associated with a road segment segment At All Other Times in section from multiple data source, wherein data sample marks on curve 210, and x-axis 210b is the time of measuring, y-axis 210a is the speed measured.In this example, shown data sample is obtained from multiple Mobile data source and one or more road traffic sensors associated with road segment segment, and in shown legend with the display of different shapes (namely, the data sample of black solid diamond " ◆ " for obtaining from road traffic sensors, and square hollow " " data sample for obtaining from Mobile data source).As described in reference to figure 2A, the shown data sample come from Mobile data source can associate with road segment segment.

Exemplary data sample comprises road traffic sensor data sample 211a-c and Mobile data source data sample 212a-d.The reported speeds of given data sample and writing time can be determined by its position on the graph.Mobile data source data sample 212d has the reported speeds of (or other speed unit) 15 mph. and is recorded at about 37 minutes (or At All Other Times unit) relative to some starting points.Will be described in more detail as following, some embodiments can be analyzed in the special time window within the shown time period or process the data obtained, such as time window 213.In this example, time window 213 comprises recorded data sample within the time interval of 10 minutes of 30 minutes to 40 minutes time.In addition, the data sample component produced in special time window can also be become two or more groups by some embodiments, such as, and group 214a and group 214b.Such as, it should be noted that shown data sample shows as dual model (bi-modal) distribution reflecting reported speeds, it has bulk data sample, is reported in the speed within the scope of 25-30 mph. of scope or 0-8 mph..May produce this dual model of speed or other multi-model (multi-modal) distribution be because, such as bottom magnitude of traffic flow pattern right and wrong are homogeneous, here due to such as make traffic stopping-traffic control signal of walking modes flowing, or road segment segment comprises multiple with the traffic lane of friction speed movement (such as, HOV track or express lane have the speed relatively higher than other non-HOV track).In this multi-model distribution that there is speed data, data sample can be divided into two or more groups to process by some embodiments, so that the process degree of accuracy that generation improves or resolution are (such as, the average velocity of each magnitude of traffic flow speed is reflected more accurately by calculating) and interested additional information is (such as, the speed of difference between HOV traffic and non-HOV traffic), or identification data sample group is got rid of (such as, not comprising the part of HOV traffic as subsequent analysis).Although do not illustrate here, this different group of data sample can identify in every way, comprises by the difference distribution modeling (such as normal or Gaussian distribution) for often organizing observation speed.

Fig. 2 C illustrates and performs to filtrator the example that the data sample eliminating that will not represent the up vehicle sailed in particular lane section was removed or considered to data sample exceptional value, it is based on the reported speeds (although the one or more parts can replaced with performing an analysis of data sample in other embodiments, no matter and be replace the speed got rid of and report) for data sample in this example.Particularly, Fig. 2 C shows table 220, and the example set which illustrates for ten data samples performs data sample exceptional value removal (quantity in actual use, being performed the data sample of analysis can be larger).Shown data sample is passable, such as, the all data samples occurred in special time window (time window 213 of such as Fig. 2 B), or alternatively can comprise special time window data sample subset (such as in the group 214a or 214b of Fig. 2 B included) or available all data samples in longer time section can be comprised.

In this example, in determined data sample group, by carrying out the velocity deviation of each speed sample in determining data sample group from the average velocity of other data sample in group, non-representational data sample is identified as the statistics exceptional value relative to other data sample.The deviation of each speed sample can be measured, the numerical value of the standard deviation such as differed relative to the average velocity of other data sample in group, its deviation ratio predetermined threshold (such as 2 standard deviations) large data sample is identified as exceptional value, and (such as, by abandoning) is got rid of from further process.

Table 220 comprises orientation row 222, which depict the content of multiple row 221a-f.The exceptional value that the often row 223a-j of table 220 illustrates for a differing data sample in ten data samples removes analysis, row 221a represents will be the data sample of often row analysis, due to each row of data sample will be analyzed, therefore it is got rid of the difference determining this result from other sample of this group.The data sample of row 223a can be referenced as the first data sample, and the data sample of row 223b can be referenced as the second data sample etc.Row 221b comprises the reported speeds of each data sample, and it is with how many mph. of measurements.Row 221c list relative to will by the data sample of given row that compares, other data sample in group, row 221d lists the speed on a rough average of the data sample group indicated by row 221c.Row 221e contains the roughly deviation between the speed and the average velocity being listed in other data sample in 221d of the data sample got rid of from row 221b, and it is with standard deviation measurement.Whether larger than 1.5 standard deviations for this example object based on the deviation listed in row 221e, whether row 221f is indicated to given data sample and should be removed.In addition, the average velocity 224 for all 10 data samples is shown as about 25.7 mph., and the standard deviation 225 of all 10 data samples is shown as about 14.2.

Like this, such as, row 223a illustrates the speed of data sample 1 is 26 mph..Next, the average velocity calculating other data sample 2-10 is about 25.7 mph..Then the deviation calculating the speed of data sample 1 and the average velocity of other data sample 2-10 is approximately .02 standard deviation.Finally, because the deviation of data sample 1 is lower than the threshold value of 1.5 standard deviations, therefore determining data sample 1 is not exceptional value.In addition, the speed that row 223c illustrates data sample 3 is 0 mph., and the average velocity of other data sample 1-2 and 4-10 is calculated as about 28.6 mph..Then the deviation calculating the speed of data sample 3 and the average velocity of other data sample 1-2 and 4-10 is approximately 2.24 standard deviations.Finally, because the deviation of data sample 3 is higher than the threshold value of 1.5 standard deviations, therefore determining data sample 3 is exceptional values.

More formally, given N number of data sample v 0, v 1, v 2..., v n, record within the given time period also associates with given road segment segment, current data sample v nto be removed, if

| v i - v i ‾ | σ i ≥ c

Wherein, v ifor the speed of analyzed current data sample; for other data sample (v 0..., v i-1, v i+1..., v n) average velocity; σ ifor the standard deviation of other data sample; C is constant threshold (such as, 1.5).In addition, as the special circumstances processing the division by 0 that may exist, if the standard deviation sigma of other data sample ibe zero and the speed of current data sample be not equal to other data sample average velocity, then remove current sample v i.

To each v iit should be noted that other data sample (v that iteration might not be wanted all 0..., v i-1, v i+1..., v n) calculate on average and standard deviation sigma i.Other data sample v 0..., v i-1, v i+1..., v naverage also can represent as follows:

v i ‾ = N v ‾ - v i N - 1

And other data sample v 0..., v i-1, v i+1..., v nstandard deviation sigma ican represent as follows:

σ i = 1 N - 2 [ ( N - 1 ) σ 2 - N ( v i - v ‾ ) 2 N - 1 ]

Wherein, N is the sum (comprising current data sample) of data sample; for all data sample v 0, v 1, v 2..., v naverage; v ibe current data sample, and σ is all data sample v 0, v 1, v 2..., v nstandard deviation.By using above-mentioned formula, can calculating mean value and standard deviation efficiently, and can calculate with Time constant particularly.Because above-mentioned algorithm is that each data sample in each road segment segment calculates mean value and standard deviation, therefore this rule runs O (MN) time, and wherein M is road hop count, and N is the data sample number of each road segment segment.

In other embodiments, also other exceptional value can be used to remove and/or data removal algorithm, can substitute or additional described rejecting outliers, such as based on neural network classifier, nature Bayes classifier, and/or regression model technology, and multiple data sample group considers the technology of (such as, if at least some data sample is not independent with other data sample) together.

Fig. 2 D illustrates the example that usage data sample performs average velocity estimation, and show be similar to described for particular lane section and the instance data sample of time period in fig. 2b.Data sample marks in curve map 230, its in x-axis 230b Measuring Time at y-axis 230a measuring speed.In certain embodiments, the average velocity of given road segment segment can periodically calculate by benchmark (such as, every 5 minutes).Each calculating can consider multiple data sample in the schedule time window (or interval) of such as 10 minutes or 15 minutes.If calculate average velocity on such time window, such as time window end or be bordering on end, then when collecting the speed of data sample, data sample in time window can weighting in every way, such as consider " age " of data sample (such as, based on to the change due to traffic, therefore the newer data sample that older data sample is unlike in closer to current time place record can provide about the such intuition of the precise information of the actual traffic situation of time window end or other current times or expection like that, and older data sample is given a discount).Similarly, in certain embodiments, other data sample attribute can be considered when weighted data sample, the type of such as data source or for data sample particular source (such as, if data sample comes from more accurate than other data source or can provide data source types or the particular source of data more better than other data source, then just heavier to its weighting), and other weighted factor type one or more.

In shown example, the average velocity for example road segment segment calculates once for every five minutes on the time window of 15 minutes.This example describes the relative weighting of two illustrated data sample 231a and 231b, because they have contribution to two each calculated average velocitys of time window 235a and 235b.Time window 235a is included in the data sample of the interocclusal record in moment 30 and 45, and time window 235b is included in the data sample of the interocclusal record in moment 35 and 50.Data sample 231a and 231b drops in time window 235a and 235b.

In shown example, each data sample weighting proportional with its age in preset time window.That is, older data sample less compared to newer data sample weight (therefore less to the contribution of average velocity).Particularly, the weight of data-oriented sample reduces according to age indication in this example.The weighted function of this decay is by illustrating corresponding to two weighting curve 232a and 232b of time window 235a and 235b respectively.Each weighting curve 232a and 232b marks data sample writing time in x-axis (level), marks weight in y-axis (vertically).The sample weights that comparatively rear (such as, closer to time window end) records in time is greater than in time the sample that comparatively early (such as, starting closer to time window) records.The weight of data-oriented sample can be found out with the place corresponding to the weight map curve intersection of interested time window to it by painting perpendicular line downwards from data sample on curve 230.Such as, weight map 232a is corresponding to time window 235a, and according to the relative age of data sample 231a (older) and 231b (newer), the weight 233a of data sample 231a is less than the weight 233b of data sample 231b.In addition, weight map 232b corresponding to time interval 235b, and can find out that the weight 234a of data sample 231a is less than the weight 234b of data sample 231b similarly.In addition, clearly, for follow-up time window, the weight of data-oriented sample decays in time.Such as, the weight 233b of the data sample 231b in time window 235a is greater than the weight 234b of the identical data sample 231b in time window 235b afterwards, because data sample 231b relatively upgrades compared to during time window 235b during time window 235a.

More normally, in one embodiment, the weight for the moment t recorded data sample of the time end relative to moment T place can represent as follows:

w(t)=e -α(T-t)

Wherein, e is known mathematics constant, and α is variable parameter (such as, 0.2).More than given, be then N number of data sample v in the time interval that moment T place terminates 0, v 1, v 2..., v nweighted mean velocity can state as follows, wherein t ifor data sample v ithe time (such as, its time be recorded) represented:

Weihgted average speed = Σ i n v i e - α ( T - t i ) Σ i n e - α ( T - t i )

And, estimate to calculate as follows to the mistake of calculated average velocity:

Errore stimate = σ N

Wherein, N is data sample number and σ is the data sample v come from average velocity 0, v 1, v 2..., v nstandard deviation.Also can be the value of the confidence that the average velocity calculating or produce determines other form in other embodiments similarly.

As will attentively, no matter to substitute or except age of data sample, data sample can based on other factors weighting.Such as, data sample can use different weighting functions (such as, the weight of data sample linearly reduces with the age instead of reduces exponentially) to carry out time weight as mentioned above but simultaneously.In addition, data sample weighting can also based on the sum of the data sample within the interested time interval.Such as, above-mentioned variable parameter α can depend on or based on data sample sum and change, so that older more at most the data sample of the quantity of data sample just produces higher punishment (such as, lower weight), the possibility of the increase of (such as, newly) data sample is more postponed to be reflected as the object that calculates average velocity.And data sample can based on comprising the other factors of data source types and weighting.Such as, it can be following situation, specific data source (such as, specific road traffic sensors, or whole traffic sensors of particular network) be all known (such as, status information based on report) or expect that (such as, based on history observation) is for unreliable or coarse.Under these circumstances, the data sample (such as, the data sample 211a of Fig. 2 B) obtained from such road traffic sensors can be fewer than the data sample weighting obtained from Mobile data source (the data sample 212a of such as Fig. 2 B).

It is the example that road segment segment performs magnitude of traffic flow estimation that Fig. 2 E simplifies based on data sample, and it such as can comprise infers the volume of traffic, density and/or occupancy.In this example, the volume of traffic of given road segment segment is expressed as in given time window in the vehicle total amount flowing through road segment segment or the vehicle total amount reached in road segment segment in time window, the traffic density of given road segment segment can be expressed as per unit distance (such as, mile or kilometer) vehicle total amount, occupation due to communication rate can be expressed as the mean time area of a room that vehicle takies particular lane section in road segment segment or point.

Given multiple will being observed runs over the different Mobile data source of given road segment segment during given time window, with the known of the total vehicle as Mobile data source or expection number percent, then can infer total volume of traffic---during time window, run over the vehicle fleet vehicle of Mobile data source (comprise be not) of road segment segment.From inferred total volume of traffic, and the average velocity of the estimation of vehicle in road segment segment, just can calculate traffic density and road occupancy further.

Estimation a kind of simple approach of total volume of traffic in particular lane section during special time window removes the quantity in the Mobile data source of this time window simply---like this, such as, if receive Mobile data sample from 25 Mobile data sources and expect that 10% of total vehicle will become Mobile data sample source in road segment segment in time window, then the total amount estimated for the time quantum of this time window is 250 actual vehicle.But due to the intrinsic changeability of vehicle arriving rate, if particularly the expectation number percent of Mobile data sample source is very little, then this approach may cause the great variety that adjacent time window total amount is estimated.Replace as one, which provide more complicated analysis, total volume of traffic of given road segment segment can be inferred as follows.The different Mobile data source of given specific quantity (such as, each portion vehicle) n, in the road segment segment of length l, in given time period τ, Bayesian statistics is used to infer the main average rate (underlying means rate) that Mobile data source arrives, λ.Can stochastic modeling, on time discrete processes corresponding to the Mobile data source that one section of road of road segment segment arrives, therefore can be described by Poisson statistics, that is:

p ( n | λ ) = λ n e - λ n !

From above formula, the possibility that n Mobile data source is observed can be calculated, given mean arrival rate λ and the vehicle number n observed.Such as, assumed average arrival rate λ=10 (vehicle/unit interval) and observe n=5 portion vehicle, then replace generation:

p ( n | λ ) = 10 5 e 10 5 ! ≈ 0.038

Represent that actual observation n=5 portion vehicle has the possibility of 3.8%.Similarly, if mean arrival rate is λ=10 (vehicle/unit interval), actual observation is 12.5% to the possibility that 10 vehicles reach (that is, n=10).

Above formula can make the possibility of the specific arrival rate λ for determining given observation n together with Bayes' theorem.As known, Bayes' theorem is:

p ( λ | n ) = p ( n | λ ) p ( λ ) p ( n )

By replacing and constant removal, can obtain as follows:

p ( λ | n ) ∝ λ n e - λ n !

From with upper, given observation n Mobile data source, can calculate the proportional or relative possibility of arrival rate λ, provides when each observed reading of given n, the probability distribution of the probable value of λ.For the particular value of n, the possibility distrabtion in each arrival rate value allows the representational arrival rate value of selection one (such as, mean value or intermediate value) and allows the degree of confidence of this value of estimation.

And, to the known percentage fixed on as total vehicle in Mobile data source on road, also as " permeability factor ", the arrival rate amount of total traffic therefore can be calculated as follows:

Total traffic volume = λ q

In certain embodiments, the total volume of traffic within the time period in road segment segment alternatively can be expressed as the total amount k flowing through the vehicle of the length l of road segment segment at time τ.

Fig. 2 E illustrates given observation sample size, the probability distribution of various total volume of traffic of given sample Mobile data source permeability factor q=0.014 (1.4%).Particularly, Fig. 2 E illustrates three-dimensional curve diagram 240, it has marked Mobile data source number (n) observed in y-axis 241, x-axis 242 denotes the traffic arrival rate amount of deduction, and in z-axis 243, denotes the possibility of the traffic value of each deduction.Such as, the graph shows given Mobile data source observation number n=0, the possibility of actual traffic amount near zero is about 0.6 (or 60%), as shown in by hurdle 244a, and time per unit actual traffic amount is about 0.1 in the possibility of 143 left and right vehicle wheels, as shown in hurdle 244b.And, given Mobile data source observation number n=28, then time per unit total actual traffic amount at 2143 left and right vehicle wheels (corresponding to about 30 the Mobile data sample sources of time per unit, the permeability factor of given example) possibility be about 0.1, as shown in hurdle 244c, which show the intermediate value close to total actual traffic amount.

In addition, can use for total traffic arrival rate amount (representing the vehicle number k arrived in the time τ of road segment segment) of the deduction of given road segment segment, the average velocity v that estimates, and average Vehicle length d calculates average occupancy and density, then

Vehicles per mile , m = k vτ

Occupancy=md

As discussed previously, the average velocity v of the vehicle in road segment segment can be obtained by operating speed estimating techniques, such as, with reference to the description that figure 2D does.

Figure 10 A-10B illustrates adjustment or revises the example of the misdata sample from the such as unreliable of road traffic sensors and obliterated data sample etc.Particularly, Figure 10 A shows the Multi-instance data readings obtained from multiple traffic sensor in each time, and it is organized in table 1000.Table 1000 comprises multiple data readings row 1004a-1004y, it is each comprises and uniquely identifies traffic sensor ID (" the identifier ") 1002a providing the traffic sensor of reading, traffic sensor data readings value 1002b comprises the traffic flow information reported by traffic sensor, the traffic sensor reading duration, 1002c reflected by the time of traffic sensor image data reading, and traffic sensor state 1002d comprises the instruction of traffic sensor mode of operation.Although traffic sensor can report the traffic flow information (such as, the volume of traffic and occupancy) of other type in other embodiments, only show velocity information in this example, and value also can with other form report.

In shown example, data readings 1004a-1004y also can be shown in table 1000 by record sheet by multiple traffic sensor collection in each time.In some cases, data readings is gathered by traffic sensor periodicity (such as, per minute, every five minutes etc.) and/or is reported by this traffic sensor with such cycle.Such as, traffic sensor 123 every five minutes image data readings, as shown in data readings 1004a-1004d and 1004f-004i, which show the multiple data readings gathered independently two days (being 8/13/06 and 8/14/06 in this example) 10: 25AM and 10: 40AM by traffic sensor 123.

Data readings 1004a-1004y shown in each comprises data readings value 1002b, and it comprises the traffic flow information observed by data transducer or obtained.Such traffic flow information can comprise travelling and arrives, to close on or by one of traffic sensor or the speed of multi-section vehicle.Such as, data readings 1004a-1004y respectively illustrates the car speed that sensor 123 arrives at four different time observations, 34 mph. (mph), 36mph, 42mph and 38mph.In addition, traffic flow information can comprise travelling and arrives, to close on or by the vehicle total amount of traffic sensor or incremental count, and no matter substitutes or except speed and/or out of Memory.Total quantity can be mounted from traffic sensor or activating, the semi-invariant of the vehicle of traffic sensor observation.Incremental count can be from sensor collection first data readings, the semi-invariant of the vehicle observed by traffic sensor.Data readings 1004w-1004x shows and has added up 316 cars and 389 cars respectively at two different timers 166.In some cases, recorded data reading can not comprise data readings value, such as, when sensor fault has appearred in given traffic sensor, thus can not gather or hourly observation or report observation (such as, due to network failure).Such as, data readings 1004k shows traffic sensor 129 can not provide data readings value in 10: the 25AM of 8/13/06 this day, as indicated in data readings value row 1002b by "--".

In addition, traffic sensor state 1002d can associate with at least some data readings, if such as traffic sensor and/or corresponding communication network provide the instruction of the mode of operation of this traffic sensor.In an illustrated embodiment, mode of operation comprises sensor function and indicates (such as normally, OK), sensor off-position (such as, OFF) instruction, sensor is handled the single value of report (such as, STUCK) indicate, and/or disconnect (COM_DOWN) with the communication link of network and indicate, as respectively at data readings 1004m, shown in 1004k, 1004o and 1004s.In other embodiments, other and/or different information of the mode of operation relating to traffic sensor can also be provided, or this operational status information can be mustn't go to.Other traffic sensor, such as traffic sensor 123 and 166 is not configured to provide traffic sensor state to indicate in this embodiment, as shown in traffic sensor status Bar 1002d "--".

Row 1004e, 1004j, 1004n, 1004q, 1004v and 1004y and row 1002e point out to record additional traffic sensor data readings in certain embodiments and/or can provide additional information and/or be recorded as the part of each data readings.Similarly, in certain embodiments, what information showed than being used in technology as described herein is few.

Figure 10 B illustrates the example of the mistake detected in the traffic sensor data readings representing the unsound traffic sensor that can not correctly work.Particularly, because a lot of traffic sensor can not provide the instruction of traffic sensor state, and the instruction due to traffic sensor state so in some cases may be insecure (such as, indication sensor function is abnormal but in fact it is normal, or indication sensor function normally but in fact it is abnormal), Using statistics and/or other technology therefore may be needed to detect unsound traffic sensor based on reported data readings value.

Such as, in certain embodiments, unsound traffic sensor can by by by the time period of given traffic sensor in certain day (such as, at 4:00PM and 7:29PM) in the current distribution of data readings reported and this sensor in the past several days (such as, past 120 days) the same time period in the historical rethinking of data readings reported compare and detect.Such distribution can produce by such as processing multiple data readings of obtaining from all traffic sensors as shown in Fig. 10 A.

Figure 10 B shows three histograms 1020,1030 and 1040, and its each expression is based on the data readings distribution of the data readings obtained from traffic sensor 123 within the interested time period.At histogram 1020, the data represented in 1030 and 1040 are dispersed to the interval of 5 mph. (such as, 0 to 4 mph., 5 to 9 mph., 10 to 14 mph. etc.) and standardization, so that the representative of every hurdle (such as hurdle 1024) occurs in the probability between this time period (such as, based at the number percent falling into data readings in the time period in this barrel) inherent 0 and 1 for the car speed of this hurdle car speed in 5 mph. of buckets (bucket).Such as, hurdle 1024 represents that the car speed between 50 and 54 mph. observed by traffic sensor 123, there is the probability of about 0.23, such as, based on the reported speeds had about 23% (containing) of the data readings obtained from traffic sensor 123 between 50 and 54 mph..In other embodiments, one or more other barrel of size can be used, and no matter except or replace the bucket of 5mph.Such as, 1mph bucket can provide thinner process interval, if but sufficient data readings can not be obtained within the time period, then also may cause the great variety between adjacent bucket, and 10mph bucket can provide less change but details is also few.In addition, although current example uses average velocity measuring as data readings analysis and comparison, other embodiments also can use one or more replacement or other except average velocity to measure.Such as, the volume of traffic and/or occupancy can be used similarly at least some embodiments.

In this example, the historical rethinking of the data readings that the Monday that histogram 1020 illustrates 120 days is in the past gathered by traffic sensor 123 between 9:00AM to 12:29PM.Histogram 1030 represents the distribution of the data readings gathered by sensor 123 between the 9:00AM to 12:29PM of specific Monday when sensor 123 function is normal.Can clearly find out, shape and the histogram 1020 of histogram 1030 are similar, assuming that similar with the travel pattern of general Monday in the travel pattern expection of specific Monday, then will discuss as following, can calculate similar degree in every way.Histogram 1040 represents the distribution of the data readings gathered by traffic sensor 123 between the 9:00AM to 12:29PM of specific Monday when sensor 123 function is abnormal, and exports the data readings that can not reflect actual traffic flow on the contrary.As obviously finding out ground, different significantly from histogram 1020 of the shape of histogram 1040, that reflects the data readings of the mistake reported by traffic sensor 123.Such as, projection huge in this distribution can be found out in hurdle 1048, and sensor 123 is stuck and reports a large amount of constant reading that can not reflect actual traffic flow when it may represent at least some between 9:00AM to 12:29PM.

In certain embodiments, although the Kullback-Leibler divergence (divergence) that can be used between two traffic sensor data distributions determines the similarity between two distributions, the similarity in other embodiments between distribution or difference also can otherwise calculate.Kullback-Leibler divergence is that the convexity of the similarity of two probability distribution P and Q is measured.It can represent as follows:

D KL ( P | | Q ) = Σ i P i log ( P i Q i )

Wherein Pi and Qi is the value (such as, each Pi and Qi is the probability that speed appears in i-th bucket) of discrete probability distribution P and Q.In shown example, be about 0.076 for the Kullback-Leibler divergence (" DKL ") 1036 of the traffic sensor of health in the data readings distribution shown in histogram 1020 and between the data readings distribution shown in histogram 1030, and be about 0.568 in the data readings distribution shown in histogram 1020 and the Kullback-Leibler divergence 1046 for unsound traffic sensor between the data readings shown in histogram 1040 distributes.As desired by possibility, DKL 1036 is significantly less than DKL 1046 (in this case, be approximately 13% of DKL 1046), that reflects histogram 1030 (such as, represent the output of the traffic sensor 123 when its function is normal) similar in appearance to histogram 1020 (such as, represent the average behavior of traffic sensor 123) be more far more than that histogram 1040 (such as, representing the traffic sensor 123 when its fault) is similar in appearance to histogram 1020.

In addition, substitute such as from Kullback-Leibler divergence come similarity or in addition, some embodiments can use other statistics to measure to detect the misdata reading provided by traffic sensor, such as statistical information entropy.The statistical entropy of probability distribution is measuring of the otherness of probability distribution.The statistical entropy of probability distribution P can represent as follows:

H ( P ) = - Σ i P i log P i

Wherein, Pi is the value (such as, each Pi is the probability that speed drops in i-th barrel of P histogram) of discrete probability distribution P.In an illustrated embodiment, the statistical entropy 1022 of the distribution shown in histogram 1020 is approximately 2.17, the statistical entropy 1032 of the distribution shown in histogram 1030 is approximately 2.14, and the statistical entropy 1042 of distribution shown in histogram 1040 is approximately 2.22.As may expectedly, statistical entropy 1042 than statistical entropy 1032 and statistical entropy 1022 all large, which reflects the traffic sensor 123 when its fault and illustrate more chaotic output mode.

In addition, the difference between two statistical entropies are measured can be measured by calculating entropy difference measurement.Can represent as follows two entropy difference measure between probability distribution P and Q:

EM=||H(P)-H(Q)|| 2

Wherein H (P) and H (Q) is respectively the entropy of probability distribution P and Q as mentioned above.In shown example, be approximately 0.0010 in the distribution shown in histogram 1020 and the entropy difference measure between the distribution shown in histogram 1030 (" EM ") 1034, and be approximately 0.0023 in the distribution shown in histogram 1020 and the entropy difference measure 1044 between the distribution shown in histogram 1040.As it is expected to ground, entropy difference measure 1044 obviously specific entropy difference measure 1034 is wanted large (large in this case twice), which reflects the statistical entropy of the distribution shown in histogram 1040 and the difference between the statistical entropy of the distribution shown in histogram 1020 wants large compared to the statistical entropy in the distribution shown in histogram 1030 and the difference between the statistical entropy of the distribution shown in histogram 1020.

Above-mentioned statistics can be used in every way to measure to detect unsound traffic sensor.In certain embodiments, the various information about the distribution of current data reading can be provided as the input to sensor health (or data readings reliability) sorter, such as, based on neural network, Bayes classifier, decision tree etc.Such as, sorter input information can comprise, such as, and the Kullback-Leibler divergence between the historical data reading distribution for this traffic sensor and the current data reading distribution for this path sensor, and the statistical entropy of current data reading distribution.Then, sorter estimates the health of this traffic sensor based on provided input, and provides the output representing healthy or unhealthy sensor.In some cases, additional information is also provided to be used as the input of sorter, the instruction of such as, time in one day (such as, time period from 5:00AM to 9:00AM), the instruction of certain day in one week or a few days (such as, from Monday to Thursday, Friday, Saturday or Sunday) and/or corresponding to certain sky in the time or one week in one day of the distribution of current and historical data reading, the size etc. of mph group.Sorter can be trained by using actual past data reading, such as comprises the expression of traffic sensor state, just as shown in Fig. 10 A.

In other embodiments, unsound traffic sensor just can be identified without the need to using sorter.Such as, if one or more statistics is measured be greater than predetermined threshold value, then can determine that traffic sensor is unsound.Such as, if the Kullback-Leibler divergence between the historical data reading distribution for traffic sensor and the current data reading distribution for this path sensor is greater than first threshold, if the statistical entropy of current data reading distribution is greater than Second Threshold, if and/or the entropy difference measure between the distribution of current data reading and the distribution of historical data reading is greater than the 3rd threshold value, then can determine that this traffic sensor is unsound.In addition, also can use other non-statistical information, whether such as traffic sensor reports can be considered to healthy or unsound sensor states.

As noted previously, although above-mentioned technology is mainly described in the context of the traffic sensor of reporting vehicle velocity information, same technology also can use other traffic flow information, comprises the volume of traffic, density and occupation rate.

Fig. 3 is the structural drawing that diagram is suitable for performing the embodiment of the computing system 300 of some of at least described technology, such as, by performing the embodiment of data sample management system.Computing system 300 comprises CPU (central processing unit) (" CPU ") 335, each I/O (" I/O ") assembly 305, storer 340 and internal memory 345, and shown I/O assembly comprises display 310, network connects 315, computer-readable medium drive 320 and other I/O equipment 330 (such as, keyboard, mouse or other printing device, microphone, loudspeaker etc.).

In an illustrated embodiment, perform various system in memory 345 to perform some of at least described technology, comprise data sample management system 350, predicted traffic information provide system 360, key road identifier system 361, road segment segment certainty annuity 362, RT information providing system 363 and other optional system of being provided by program 369, these various executive systems are all referred to as traffic information system here usually.Computing system 300 and its executive system can via network 380 (such as, internet, one or more mobile telephone networks etc.) communicate with other computing system, such as each client device 382, based on the client of vehicle and/or data source 384, road traffic sensors 386, other data source 388 and third party's computing system 390.

Particularly, data sample management system 350 obtains the information of the various situation data about current traffic condition and/or previous observations from each source, such as from road traffic sensors 386, based on vehicle Mobile data source 384 and/or other moves or non-moving data source 388 obtains.Then data sample management system 350 is by filtering (such as, consider to remove data sample) and/or adjustment is (such as, error recovery) data that obtain of the data use that is other assembly and/or system and preparing, then use the data prepared such as, to estimate the road traffic condition of each bar road segment segment, the magnitude of traffic flow and/or speed.In embodiment shown in this, data sample management system 350 comprises data sample filterer component 352, sensing data adjustment assembly 353, data sample exceptional value removes assembly 354, data sample speed assessor component 356, data sample flow estimation assembly 358 and optional sensor data collection assembly 355, wherein assembly 352-358 perform and be similar to before the function described by corresponding assembly in FIG (such as, data sample filterer component 104, sensing data adjustment assembly 105, data sample exceptional value removes assembly 106, data sample speed assessor component 107, data sample flow estimation assembly 108 and optional sensor data collection assembly 110).In addition, in at least some embodiments, data sample management system performs the estimation of road traffic condition in mode when in real time basic or approximate real, such as, within a few minutes, obtain bottom data (himself can obtain from data source in substantially real-time mode).

Other traffic information system 360-363 and 369 and/or third party's computing system 390 then can use the data provided by data sample management system in every way.Such as, predicted traffic information provides system 360 can obtain (directly, or indirectly via database or memory device) this prepared data to be to produce further traffic condition predictions at multiple future time, and information of forecasting is supplied to other receiving end one or more, such as other traffic information system one or more, client device 382, based on client 384 and/or third party's computing system 390 of vehicle.In addition, RT information providing system 363 can obtain information about estimated road traffic condition from data sample management system, and by road traffic condition information with in real time or be bordering on real-time mode and be supplied to its side (such as, client device 382, client 384 and/or third party's computing system 390 based on vehicle)---when data sample management system also with this in real time or be bordering on real-time mode perform estimation time, the take over party of data that comes from RT information providing system can browse based on the actual vehicle travel conditions of the same period on one or more road segment segment and use about the current traffic condition in these road segment segment information (as by the Mobile data source travelled in these road segment segment and/or sensor reported, and other data source provides the information about the actual vehicle travel conditions in these road segment segment).

Client device 382 can adopt various forms in various embodiments, and usually can comprise any communication facilities with other request of producing to traffic information system and/or any computing equipment receiving information from traffic information system.In some cases, client device can perform the operable interactive controlling application program of user (such as, Web browser) to produce to the request of the information relating to traffic (such as, the future traffic condition information of prediction, in real time or be bordering on real-time current traffic condition information etc.), and in other situations, the information of traffic that what at least some was such relate to can be off-loaded automatically from one or more traffic information system and be sent to client device (such as, text message, new Web page, specific routine data renewal etc.).

Road traffic sensors 386 comprise multiple be arranged on such as one or more geographic area each street, in expressway or other roads, upper or neighbouring sensor.These sensors can comprise loop sensor, can measure time per unit by the quantity of the vehicle of these sensors, the speed of vehicle and/or other data relating to the magnitude of traffic flow.In addition, such sensor can comprise camera, motion sensor, radar ranging equipment, based on RFID equipment and to be located close to or near the sensor of other type of road.Road traffic sensors 386 can periodically or continually by based on wired or by using the network 380 of one or more data exchange mechanism (such as, push away, draw, token, request-response, point-to-point etc.), the data readings of measurement is supplied to data sample management system 350 based on wireless data link.In addition, although do not illustrate here, but in certain embodiments, one or more gatherers (such as, the governmental transportation body of operation sensor) of such road traffic sensors information can replace to obtain raw data and make data concerning traffic information system being available (no matter be original form or after it is processed).

Other data source 388 comprises other data source polytype, its information that can make for providing to user, consumer and/or other computing system about traffic by one or more traffic information system.Such data source comprises the Map Services and/or database that can provide about road network information, such as each bar road connectedness each other and relate to the traffic control signal (such as, the existence in traffic control signal and/or speed limit district and position) of such road.Other data source can also comprise the source of relative effect and/or the reflection event of traffic and/or the information of situation, such as short-term and long-range weather forecasting, school's schedule and/or calendar, schedule of events and/or calendar, the traffic accident report, road work information, dealing with holidays etc. that are provided by manual operation person (such as, the first present members, law enfrocement official, highway crews, news media, tourist etc.).

The clients/data sources 384 based on vehicle is in this embodiment each can be positioned at vehicle data to be supplied to one or more traffic information system and/or from the computing system of these system receives data one or more and/or communication system.In certain embodiments, data sample management system 350 can use the Mobile data source based on vehicle and/or other distributed network based on the Mobile data source (not shown) of user that provide the information relating to current traffic condition for the use of traffic information system.Such as, every portion vehicle or other Mobile data source can have GPS (" GPS "), and equipment (such as, there are the mobile phone of GPS function, independently GPS device etc.) and/or other can determine the geolocation device in geographic position, and may out of Memory be also had, such as speed, direction, height above sea level and/or other relate to vehicle travel data, and geolocation device or other disparate communications apparatus sometimes obtain and provide such data to one or more traffic information system (such as, passing through wireless link).Such Mobile data source will discuss in more detail elsewhere.

Alternatively, based on some or all of the clients/data sources 384 of vehicle each can have be positioned at vehicle computing system and/or communication system to obtain information from one or more traffic information system, such as, in order to the use of vehicle driver.Such as, vehicle can comprise embedded panel board (in-dash) navigational system of Web browser or other controlling application program with installation, user can use this system to ask traffic-relevant information from one of traffic information system (such as predicted traffic information provides system and/or RT information providing system), or these requests can be sent by the portable set of the user in vehicle.In addition, one or more traffic information system automatically can will relate to the information transmission of traffic to such client device based on vehicle based on the reception of lastest imformation or generation.

Third party's computing system 390 comprises one or more optional computing system, and it is by such as receiving about a side of the data of traffic and the operator of other people instead of the traffic information system of the side etc. of usage data in some way operate from one or more traffic information system.Such as, third party's computing system 390 can be such system, it is from one or more traffic information system receiving traffic information, and involved information (no matter being received information or the out of Memory based on received information) is supplied to user or other people (such as, by Web entrance or subscription service).Alternatively, third party's computing system 390 can be operated by a side of other type, such as collect and the media organization of report predicted traffic condition to consumer, or to the Online Map company that their user provides the information about traffic to be used as a travel-planning services part.

As above will attentively, the data that predicted traffic information provides system 360 to use to be prepared by data sample management system 350 and other assembly are in the embodiment shown to produce the traffic condition predictions in future of multiple future time.In certain embodiments, the generation of forecast employs probabilistic technique, this incorporates each that various types of input data think many road segment segment and repeatedly produce the forecast of a series of future time, such as based on the road network in given geographic area changing the present situation and in real time.And, in at least some embodiments, for using in the prediction of the future traffic condition of each interested geographic area and automatically create one or more predictability Bayes or other model (such as giving, decision tree), such as, based on the historical traffic conditions observed of these geographic areas.The future traffic condition information of predictability can use to help travelling or other object in every way, so that based on the prediction plan of the traffic of the road of multiple future time by the optimal route of road network.

And, road segment segment certainty annuity 362 can use provide the Map Services of the information relating to road network in one or more geographic area and/or database with automatically determine and manage relate to may the information of electronic map that uses by other traffic information system.The information of such electronic map can comprise will by the determination of the specific part of the road as interested road segment segment (such as, based on the traffic of these road sections with other adjacent road part), and in the road segment segment of given road network and the instruction of interested out of Memory (such as, the physical location of road traffic sensors, case point, terrestrial reference; About function road class and other information about traffic characteristic; Deng) between the association that automatically produces or relation.In certain embodiments, road segment segment certainty annuity 362 periodically can perform and in storer 340 or database (not shown), store the information of its generation in order to the use of other traffic information system.

In addition, key road identifier system 361 uses the road network representing given geographic area and the traffic related information for that geographic area, thinking and follow the tracks of and estimate road traffic condition and automatically identify interested road, such as, is the use of other traffic information system and/or traffic data clients.In certain embodiments, the automatic identification of interested road (or one or more road segment segment of road) can at least in part based on following factor, the such as value of the peak value volume of traffic or other flow, the value of peak value traffic congestion, the change on the same day of the volume of traffic or other flow, the change on the same day of congestion in road, (inter-day) in the daytime change of the volume of traffic or other flow, and/or the Daytime varieties of congestion in road.Such factor can be analyzed by such as primary clustering (principal component) analysis, such as by first calculating the covariance matrix S for the traffic related information of all roads (or road segment segment) in given geographic area, then calculate the eigen decomposition of covariance matrix S.Then, in the descending of eigenwert, the eigenvector of S represents the combination of the road (or road segment segment) independently the change of observed traffic being had to the strongest contribution.

In addition, Real-time Traffic Information provides or presents system by RT information providing system, or alternatively can be provided by other program 369 one or more.Information providing system can use data that to be analyzed by data sample management system 350 and/or other assembly (such as predicted traffic information provides system 360) and provide for operation or use client device 382, provides traffic-information service based on the consumer of the client 384, third party's computing system 390 etc. of vehicle and/or commercial entity, so as at least in part based on the data sample obtained from vehicle and other Mobile data source with in real time or be bordering on real-time mode and provide data.

Can predict, shown computing system is only schematic and does not attempt to limit the scope of the invention.Computing system 300 can with other unshowned equipment connection, comprise by the network of one or more such as internet or via Web.In general, " client " or " server " computing system or equipment, or traffic information system and/or assembly, can comprise can be mutual and perform the combination in any of hardware and software of described type of functionality, include but not limited to desktop or other computing machine, database server, the network storage equipment and other network equipment, PDA, cellular mobile phone, wireless telephone, beeper, communicator, internet application, based on TV system (such as, use Set Top Box and/or individual/digital video recorder) and comprise other consumer products various with suitable interactive communication ability.In addition, the function provided by shown system component in certain embodiments can be integrated in less assembly or be distributed in additional assembly.Similarly, the function of some in shown in certain embodiments assembly can not be provided and/or can obtain other additional function.

In addition, although various project is as directed can be stored in storer or memory storage while use, in order to the object of memory management and/or data integrity, these projects or their part can be transmitted between storer and other memory device.Alternatively, some or all of component software and/or module in other embodiments can be performed in the storer on another equipment and be communicated with shown computing system by the communication of intercomputer.Some or all of system component or data structure also can be stored in computer-readable medium (such as, as software instruction or structural data), such as, by suitable driver or the hard disk read by suitable connection, storer, network or portable media article.System component and data structure also can be transmitted as produced data-signal (such as on various computer-readable transmission medium, a part as carrier wave or other analog or digital transmitting signal), comprise based on wireless and based on the medium of wired/cable, and various forms can be adopted (such as, as a part that is single or multiplexed analog signal, or as multiple discrete digital packets or frame).In other embodiments, such computer program can also adopt other form.Therefore, the present invention also can realize with other Computer Systems Organization.

Fig. 4 is the process flow diagram of the exemplary embodiment of data sample filterer device routine 400.This routine can be provided by the execution of the embodiment of the data sample filterer component 104 of the data sample filterer component 352 of such as Fig. 3 and/or Fig. 1, to receive the data sample corresponding to road in geographic area, and filter out uninterested data sample for estimation below.The data sample filtered then can use in every way subsequently, such as, use the data sample of filtration to calculate the average velocity in interested particular lane section and to calculate other feature about the magnitude of traffic flow for such road segment segment.

Routine starts in step 405, is that the geographic area of special time period receives data sample group here.In step 410, routine is then alternatively based on some or all generation additional informations that other relevant data sample is these data samples.Such as, if the particular data sample for vehicle or other Mobile data source lacks interested information (speed in such as Mobile data source and/or orientation or direction), then such information can in conjunction with to identical Mobile data source previously and subsequent data sample one or both of and determining.In addition, in at least some embodiments, the information for specific Mobile data source can collected from multiple data sample estimates the additional information type about this data source, to estimate the behavior of the data source within the time period across multiple data sample (such as, determine whether that vehicle has stopped a few minutes instead of temporarily stop to be used as one or two minute the normal wagon flow of traffic, such as, meet stop sign or stop light).

After step 410, although routine proceeds to step 415 to attempt each data sample to associate with the particular lane section of this road with the road in this geographic area, but this step can not be performed or otherwise perform in other embodiments, if such as at least the initial association of data sample and road and/or road segment segment receives in step 405, if or alternatively whole routine performs one time for a road segment segment, thus all data samples received in step 405 come corresponding to a road segment segment as a group.In an illustrated embodiment, data sample can perform with associating of road segment segment in every way with road, such as independent geographic position based on associating with this data sample and carry out initial association (such as, being associated with road segment segment with nearest road by data sample).And, this association can comprise additional analysis alternatively with concise or revision initial association---such as, if location-based analysis instruction has multiple possible road segment segment for data sample, (such as many road segment segment are used for a specific road, or alternatively many road segment segment are used for closing on but incoherent road), then such analyzing adjuncts can use such as speed to affect with the out of Memory in direction and associate (such as, by merging positional information and other such factor one or more in the mode of weighting).Like this, such as, if the reported position of data sample is between expressway and adjacent frontage road, the information of institute's reported speeds of relevant data sample then just can be used to help be associated with suitable road by this data sample (such as, by determining can not come from the frontage road with 25 mph. of speed limits with the data sample of the velocity correlation of 70 mph.).In addition, in the certain extension of road or other road sections and many different road segment segment (such as, for the road of two-way traffic, wherein traveling is in one direction modeled as first section and traveling is in the other directions modeled as different second section, or alternatively for the expressway of multilane, HOV track is modeled as the non-HOV track independently road segment segment adjacent with one or more) when being associated, the additional information of the such as relevant data such as speed and/or direction sample can be used select the road segment segment for road most possible this data sample.

After step 415, routine proceeds to step 420 and thinks that follow-up process filters out any data sample do not associated with interested road segment segment, comprises the data sample (if there is) do not associated with any road segment segment.Such as, the part of specified link or road may not be interested to subsequent analysis, such as get rid of the road of specific function road class (such as, if the size of road and/or the volume of traffic be not large enough to can have interested), or due to such as ramp, expressway or special road or cross/traffic characteristic of such road sections such as point cross road can not reflect expressway as a whole, therefore get rid of such road sections.Similarly, when many road segment segment associate with the specific part on road, some road segment segment may interested to some object, if the behavior such as only having non-HOV track is interested to specific purpose, if or the track of both direction only has a direction to be interested, then for HOV track is got rid of in expressway.Although after step 420, routine proceeds to step 425 to determine whether the behavior filtering data sample based on data source, and filtration so in other embodiments also can not be performed or also can perform always.In an illustrated embodiment, if perform filtration based on the behavior in source, then routine proceeds to step 430 to perform such filtration, such as remove corresponding to its behavior can not reflect want measured interested to magnitude of traffic flow behavior data source data sample (such as, get rid of engine start in time expand section and the vehicle stopped, get rid of within the time period extended at the vehicle etc. spinned in ground or parking lot or other zonule that stops).After step 430, if or alternatively determine in step 425 not filter based on the behavior of data source, then routine proceeds to the data that step 490 thinks follow-up use stored filter, but the data of filtering in other embodiments alternatively can directly be supplied to one or more client.Then routine proceeds to step 495 to determine whether to continue.If continued, then routine turns back to step 405, if do not continued, then arrives step 499 and terminates.

Fig. 5 is the process flow diagram of the exemplary embodiment of data sample exceptional value remover routine 500.This routine can be removed assembly 354 and/or Fig. 1 data sample exceptional value by the data sample exceptional value such as performing Fig. 3 is removed the embodiment of assembly 106 and provide, thus removes the data sample for this road segment segment being exceptional value relative to other data sample of road segment segment.

This routine starts in step 505, receives wherein and is used for road segment segment and one group of data sample of time period.The data sample received can be, the data sample of the filtration such as obtained from the output of data sample filterer device routine.In step 510, data sample is then divided into multiple groups to reflect road segment segment dissimilar parts and/or different behavior by routine alternatively.Such as, if track, many expressways be included in together as single road section a part and these many tracks comprise at least one HOV track and one or more non-HOV track, if the magnitude of traffic flow then within the time period is significantly different between HOV and non-HOV track, then the vehicle on HOV track can with the Vehicles separation on other track.Can perform such grouping in every way, such as, data sample be fitted to many curves, every bar curve represents the typical data sample changed (such as, normal state or Gaussian curve) in particular data sample group.In other embodiments, also such grouping can not be performed, if such as alternatively split road segment segment all reflect similar behavior (such as, if the expressway with HOV track and other non-HOV track is alternatively split into many road segment segment) for use in all data samples of this road segment segment.

Routine proceeds to step 515, is each (if do not perform the separation of the data sample of step 510, then all data samples are regarded as a group) of one or more data sample group, calculates the average traffic condition characteristic of all data samples.This average traffic condition characteristic can comprise, such as, and average velocity, and such as relative to the corresponding statistical information of the standard deviation etc. of intermediate value.Routine then proceeds to step 520, each to this one or more data sample group, performs removal one (leave-one-out) continuously and analyzes to select the target data sample that specifically will be temporarily removed and be remaining traffic feature determination average traffic condition characteristic.The average traffic condition characteristic for remaining data sample and from step 515 for all data samples average traffic condition characteristic between difference larger, then removed target data sample is that the possibility of the exceptional value of the public characteristic that can not reflect other remaining data sample is larger.In step 525, routine then performs the outlier detection of one or more addition type alternatively, thus remove continuously the group of two or more target data sample thus estimate their joint effect, but also can not perform so additional outlier detection in certain embodiments.After step 522, routine proceeds to step 590 and in step 520 and/or 525, is identified as the data sample of exceptional value to remove, and is follow-up use and store remaining data sample.In other embodiments, remaining data sample alternatively can be transmitted to one or more client and use by routine.Routine then arrives step 595 to determine whether to continue.If continued, then routine turns back to step 505, if do not continued, then routine proceeds to step 599 and terminates.

Fig. 6 is the process flow diagram of the exemplary embodiment of data sample speed estimator routine 600.This routine can be provided by the data sample speed assessor component 107 of the data sample speed assessor component 356 and/or Fig. 1 that perform such as Fig. 3, such as, within the time period, estimate the current average velocity of this road segment segment based on each data sample for road segment segment.In the exemplified embodiment, routine is the Continuous plus of each execution road segment segment average velocity of multiple time interval or time window within the time period, but each the calling of routine alternatively can for the single time interval (such as, via multiple routine call estimation multiple time interval) in other embodiments.Such as, if the time period is 30 minutes, new average velocity then within every five minutes, can be performed calculate, such as with the time interval of 5 minutes (and therefore each time interval not overlapping with the previous or follow-up time interval), or with the time interval of 10 minutes (time interval therefore with adjacent is overlapping).

This routine starts in step 605, receive instruction, the data sample of its instruction road segment segment within the time period (such as, data sample from the data readings of Mobile data source and physical sensors is come), or the insufficient data of instruction road segment segment within the time period, but a data sample can only be received from Mobile data source and sensor data readings in certain embodiments.The data sample received can be, such as, obtains from the output of data sample exceptional value remover routine.Similarly, the instruction of inadequate data can be obtained from data sample exceptional value remover routine.In some cases, the instruction of inadequate data can based on having data sample in shortage, such as wrong (such as, adjusting assembly 105 by the sensing data of Fig. 1) when not carrying out data sample from the Mobile data source associated with road segment segment and/or lose when some or all data readings of road segment segment or be detected as within the time period.In this example, routine continues to determine whether to have received the inadequate instruction of data in step 610.If so, then routine proceeds to step 615, if not, then routine proceeds to step 625.

In step 615, routine performs the embodiment (describing with reference to Figure 14) of magnitude of traffic flow estimation device routine to obtain the average traffic speed of the estimation of road segment segment in the time period.In step 620, routine then provides the instruction of the average velocity of estimation.In step 625, routine starts from first time interval and is want estimated average velocity to select the next time interval or time window.In step 630, routine is then the average traffic speed of data sample calculating weighting within this time interval, and based on one or more factor to data sample weighting.Such as, in an illustrated embodiment, change (such as to the weighting of each data sample based on the stand-by period of data sample, with linear, index, or step-wise fashion), such as give near the larger weight (because they more can be reflected in the actual average speed of time interval end) of the data sample of time interval end.In addition, data sample can also the weighting based on the source of data further in an illustrated embodiment, such as, no matter lay particular stress on or on the low side, is different from the data sample weighting to coming from vehicle and other Mobile data source to the data readings weighting come from physical sensors.In addition, in other embodiments, various other factors can be used in weighting, comprise based on each sample---such as, the data readings weighting to coming from another physical sensors can be different to the data readings weighting come from a physical sensors, thus the available information of reflection associated sensor (such as, one in physical sensors is intermittent error or has more coarse data readings resolution than another sensor), and the data sample come from a vehicle or other Mobile data source can similarly based on about Mobile data source information with the data sample differently weighting come from another such vehicle or Mobile data source.Other type of the factor that can use in weighting in certain embodiments comprises the value of the confidence or other estimation of possible errors in particular data sample, the degree that particular data should associate with particular lane section etc.

After step 630, routine proceeds to step 635 to provide the instruction of average computation traffic speed in the time interval, such as, store this information for follow-up use and/or will be supplied to client to information.In step 640, routine obtains obtainable additional data sample after time period inherent step 605 receives information subsequently alternatively.Then determine whether will calculate the more time interval in step 645 within the time period, and if like this, then routine turns back to step 625.If alternatively do not have the more time interval, or after step 620, then routine proceeds to step 695 to determine whether to continue.If continued, then routine turns back to step 605, and if not, then proceeds to step 699 and terminates.

Fig. 7 is the process flow diagram of the exemplary embodiment of data sample flow estimation device routine 700.Routine can be passed through, such as, perform the embodiment of the data sample flow estimation assembly 358 of Fig. 3 and/or the data sample flow estimation assembly 108 of Fig. 1 and provide, to estimate traffic traffic characteristic instead of the average velocity in particular lane section in special time period.In the exemplified embodiment, the vehicle total amount (or other Mobile data source) wanting estimated traffic characteristic be included in time period inherent particular lane section to arrive or exists, and within the time period road segment segment percentage occupancy with reflect the point of road segment segment or region the percentage of time that covers by vehicle.

Routine starts in step 705, receives instruction wherein, its instruction time section road segment segment data sample and within the time period average velocity of road segment segment, or not enough data of road segment segment in the time period.Data sample can be from, and such as, the output of data sample exceptional value remover routine obtains, and average velocity can be from, and the output of such as data sample speed estimator routine obtains.The instruction of not enough data can be from, and the output of such as data sample exceptional value remover routine obtains.In some cases, the instruction of not enough data can based on having data sample in shortage, such as when not carrying out data sample from the Mobile data source associated with road segment segment maybe when some or all sensor data readings for road segment segment are lost or to be detected as be wrong (such as, adjusting assembly 105 by the sensing data of Fig. 1) within the time period.Routine then continues to determine whether to have received the instruction of not enough data in step 706.If so, then routine proceeds to step 750, if not, then routine proceeds to step 710.

In step 750, routine performs the embodiment of magnitude of traffic flow estimation device routine (describing with reference to Figure 14) to obtain total amount and the occupancy of the estimation of road segment segment within the time period.In step 755, routine then provides the total amount of estimation and the instruction of occupancy.

In step 710, routine determines the vehicle number (or other Mobile data source) providing data sample, such as, by being associated with specific Mobile data source by each data sample.In step 720, routine then determines the arrival rate providing the road segment segment of the vehicle of this data sample most possible based on determined vehicle number in probability upper part.In certain embodiments, probability determines the also further information used about the prior probability of such vehicle fleet size and the prior probability of specific arrival rate.In step 730, routine is then such as based on the determined quantity of vehicle with infer by the sum of all vehicles of road segment segment within the time period about the number percent providing the vehicle of data sample to account for vehicle fleet, and the further fiducial interval of total amount inferred of estimation.In step 740, routine then infers the percentage occupancy of road segment segment within the time period based on inferred total amount, average velocity and average Vehicle length.Also the magnitude of traffic flow feature of other interested type can be estimated in other embodiments similarly.In an illustrated embodiment, routine then proceeds to step 790 with the instruction of the percentage occupancy of the total amount and deduction that provide deduction.After step 755 or 790, if determine to continue in step 795; Then routine turns back to step 705; If do not continued, then proceed to step 799 and terminate.

Figure 11 is the exemplary embodiment of sensor data readings error detector routine 1100.Routine can be by, and such as, the sensing data adjustment assembly 353 and/or Fig. 1 sensing data adjustment assembly 105 that perform Fig. 3 provide, thus determine the health of one or more traffic sensor.In the exemplified embodiment, based on the traffic sensor reading recently obtained within the indicated time period, perform this routine to determine the health of one or more traffic sensor each time of one day.In addition, in various embodiments, the data that the traffic for each type one or more is measured and exported by traffic sensor can by this routine analyses, such as traffic speed, quantity, occupancy etc.And, the data of some at least traffic can be measured in every way and/or collect, such as with various spaced horizontal (such as, 5mph bucket for the data group of velocity information), and this routine can with each (or other combined horizontal) of one or more each spaced horizontal measured for one or more traffic for specific traffic sensor analyzes data in certain embodiments.

This routine starts in step 1105, and receive one or more traffic sensor and selected time classification (such as nearest time classification, if routine performs and provides result to be bordering on real-time mode after each time classification, or one or more previous time classification selected for analysis) instruction, but alternatively can be instructed to multiple chronological classifications in other embodiments.In certain embodiments, time eachly can comprise time point classification (such as, 12:00AM-5:29AM and 7:30PM-11:59PM, 5:30AM-8:59AM by it, 9:00AM-12:29PM, 12:30PM-3:59PM, 4:00PM-7:29PM, and 12:00AM-11:59PM) and/or date category is (such as, Monday is to Thursday, Friday, Saturday and Sunday, or alternatively there is Saturday together with Sunday in groups) time classification and modeling.Specific chronological classification can be selected in every way in various embodiments, comprise traffic during being reflected in it and expect to have the time period of similar characteristics (such as, based on call duration time and pattern, or the consistent behavior of other reflection traffic), if such as traffic between the lights with relative rarity in early morning, then they are formed one group together.In addition, in certain embodiments, the time period with similar magnitude of traffic flow feature can be determined by analysis of history data, thus no matter distinguish different traffic sensors (such as, by geographic area, road, single-sensor etc.) with artificial or automatic way selection time classification.

In step 1110 is to 1150, routine then performs circulation, and it is analyzed from each next traffic sensor data readings for the one or more traffic sensors indicated by indicated time classification to determine the traffic sensor health status of each traffic sensor during this time classification wherein.In step 1110, from first traffic sensor, next traffic sensor in one or more traffic sensors indicated by routine selection, and select indicated by time classification (or, if alternatively indicate multiple time classification in step 1105, be then the next one combination of traffic sensor and indicated time classification).In step 1115, routine average historical data reading retrieved for traffic sensor in selected time classification distributes.In certain embodiments, the distribution of historical data reading can based on the data readings provided in selected time classification by traffic sensor (such as, striden across such as nearest 120 days or time period of prolongation in recent 120 day cycle etc. includes between 4:00PM and 7:29PM on Monday to the date of Thursday).

In step 1120, routine continues as selected traffic sensor and the distribution of selected time classification determination target traffic sensor data.In step 1125, routine then determines the similarity of the distribution of target traffic sensor data readings and the distribution of historical traffic sensor data readings.As other places more detailed description, in certain embodiments, can be determined by the Kullback-Leibler divergence calculated between the distribution of target traffic sensor data readings and the distribution of historical traffic sensor data readings in the tolerance of such similarity.In step 1130, as more discussed in detail elsewhere, routine then determines the information entropy that target traffic sensor data readings distributes.

In step 1135, routine then by use various information be select time classification estimation the health of traffic sensor select classify (such as to perform health, instruction " health " or " unhealthy ", or the value on " health " yardstick, such as from 1 to 100), its comprise in this embodiment determined similarity, determined entropy and the time classification selected (such as, moment at the hour classification selected, such as 4:00PM to 7:29PM, and/or the date category selected, such as Monday is to Thursday).In other embodiments, out of Memory type can be used, such as, want the instruction (such as, for the 5mph bucket of the data group of velocity information) of the interval degree of measured data.In one embodiment, neural network can be used classify, and in other embodiments, other sorting technique can be used, comprise decision tree, Bayes classifier etc.

In step 1140, routine then and/or other factors healthy based on the traffic sensor of estimation for select traffic sensor and the time classification determination traffic sensor health status (being health or unhealthy in this example) select.In certain embodiments, the health no matter when for the traffic sensor of selected time classification is estimated as health in step 1135, then can be considered to healthy for the health status of traffic sensor.In addition, no matter when for the health of the traffic sensor of selected time classification be estimated as unhealthy (such as, in step 1135), and moment at the hour classification of the related covering of time classification tool selected enough large time period (such as at least 12 or 24 hours), then the health status for traffic sensor can be thought unsound.And, in certain embodiments, can retrieve and use the information about relating to time classification (such as one or more previous and/or follow-up time period), thus within the longer time period (such as, one day), the health of traffic sensor be classified.Such logic reduces the interim abnormal traffic pattern accurately reported based on sensor and carries out the negative risk determining (such as determining that the health status of traffic sensor is unsound when in fact traffic sensor is healthy) of mistake to sensor health status.

Such as, due to because of external factor (such as, traffic hazard, weather accident etc.) in data readings change on significant same day may produce the negative of mistake and determine.Such as, the traffic accident of special traffic sensor place or generation near it may cause traffic sensor to provide abnormal and irregular data readings within the relatively short time period (such as, one to two hours).If the determination of sensor health status is only based on the main data readings obtained in the distribution time caused by traffic hazard, then what just likely lead to errors negatively determines.By based on the data readings obtained from the relatively large time period (such as, 12 or 24 hours) to determine the state of unsound sensor, the negative risk determined of such mistake can be reduced.On the other hand, mistake negative is determined (such as determining that the health status of traffic sensor is unsound when in fact traffic sensor is healthy), and in general possibility is very low, because the traffic sensor of fault can not provide the data readings being similar to historical data reading (such as, reflecting general travel pattern).Similarly, can suitably determine that the health status of traffic sensor is healthy based on the relatively short time period.

Some embodiments can by reflect the routine of the time classification of short period section shown in every day multiple exercise (such as, to have the chronological classification of moment at the hour classification extending first first three hour, within every three hours, perform a routine) and to reflect that the time classification on whole previous date at least performs a routine (such as in every day, with the time classification of moment at the hour classification extending previous 24 hours, at executive routine at midnight) and realize this different logic.

In addition, the determination of sensor states can based on other factors, such as whether for selected time classification obtains the data readings of sufficient amount (such as, because sensor is report data reading off and on) and/or based on the instruction (such as, traffic sensor is stuck) of the sensor states provided by traffic sensor.

In step 1145, routine provides the health status of determined traffic sensor.In certain embodiments, can be by other assembly (such as, the sensor data collection assembly 110 of Fig. 1) follow-up use and store traffic sensor health status (such as, be stored in database or file system) and/or it is directly supplied to other assembly (such as, data sample exceptional value removes assembly).In step 1150, routine determines whether there is more sensor (or combination of traffic sensor and time classification) and will process.If so, then routine proceeds to step 1110 and continues, if not, then proceed to step 1155 to perform other suitable action.Other action this can comprise, such as, for each one or more time classifications for multiple traffic sensor each come periodically (such as, once a day, inferior on every Mondays) repeatedly calculate historical data reading distribution (such as, at least 120 days).By periodically repeatedly calculating the distribution of historical data reading, in the face of the traffic gradually changed, routine can continue the determination providing accurate traffic sensor health status.After step 1155, routine proceeds to step 1199 and returns.

Figure 12 is the process flow diagram of the exemplary embodiment of data readings error corrector routine 1200.This routine can be passed through such as, and the sensing data of the sensing data adjustment assembly 353 and/or Fig. 1 that perform Fig. 3 adjusts assembly 105 and provides, thus determines the data readings after the correction of one or more traffic sensor associated with road segment segment.In the illustrated exemplary embodiment, this routine can periodically be performed (such as every five minutes) to correct the data readings for being identified as unsound traffic sensor by sensor data readings error corrector routine.In other embodiments, this routine can be performed as required, such as, by sensor data collection device routine, to obtain the data readings after for the correction in particular lane section, or alternatively can not be used in various environment.Such as, in general by determining all data samples in particular lane section (such as, come from multiple data source, such as can comprise the polytype in the Mobile data source of traffic sensor and one or more dissimilar types) whether provide enough data to analyze the traffic flow conditions of this road segment segment to perform analysis and the correction of data, if not, then do not perform the correction of the data come from each traffic sensor.

This routine starts in step 1205, wherein it receives the instruction of the road segment segment associated with one or more traffic sensor (such as, unsound result has been classified as) by that come from sensor data readings error detector routine, one or more associated traffic sensor, and receive the instruction of one or more time classifications (such as, wherein being classified as be unsound time classification at least potentially at least one of traffic sensor of association) that will be processed alternatively.In other embodiments, interested one or more traffic sensors can otherwise indicate, such as, by directly receiving the instruction of one or more traffic sensor.In step 1210 to 1235, routine performs a circulation, wherein its process unsound traffic sensor in indicated road segment segment, with during one or more time classification (such as in step 1205 indicated time classification) determine and data readings after correction is provided for these traffic sensors.

In step 1210, from first, routine selects the unsound traffic sensor of the next one in indicated road segment segment.This routine also by select one or more during it traffic sensor be previously designated as the time classification that unsound time classification etc. selects to use, such as one or more time classification indicated in step 1205.In step 1215, routine determines the traffic sensor that also can be used to assist the reading for the unhealthy sensor of selected time classification to correct whether in indicated road segment segment with other enough health.This is determined can whether based on existing in the instruction road segment segment during selected time classification at least predetermined amount of data be (such as, at least two) and/or predetermined percentage is (such as, at least 30%) healthy traffic sensor, and it is also conceivable to the relative position (such as, adjacent or neighbouring traffic sensor can be better than the sensor away from unhealthy traffic sensor) of healthy traffic sensor in the road segment segment of instruction.If determine to there is enough healthy traffic sensors in step 1215, then routine proceeds to step 1220, determines the correction data reading of unhealthy traffic sensor here based on the data readings come from other healthy traffic sensor in the road segment segment for selected time classification.Correction data reading can be determined in every way, such as, by calculating the average of the two or more data readings obtained from the healthy traffic sensor the instruction road segment segment in institute's classification seclected time.In certain embodiments, all healthy traffic sensors may be used to averaging, but can only use selected healthy traffic sensor in other embodiments.Such as, if the predetermined percentage of the traffic sensor in indicated road segment segment (such as, at least 30%) be healthy during selected time classification, all healthy traffic sensors then can be used to carry out averaging, otherwise the healthy traffic sensor of nearest predetermined quantity (such as, at least two) can only be used.

If alternatively determine there is no enough healthy traffic sensors in step 1215 in the instruction road segment segment for classification seclected time, then routine proceeds to step 1225, and here it attempts the correction data reading determining unhealthy traffic sensor based on the out of Memory relating to this traffic sensor/or road segment segment.Such as, such information can comprise the predict traffic conditions information for road segment segment and/or unhealthy traffic sensor, for the forecast traffic related information of road segment segment and/or unhealthy traffic sensor, and/or for the history average traffic condition information of road segment segment and/or unhealthy traffic sensor.Various logic can be performed to reflect the relative reliability of various information type.Such as, in certain embodiments, usage forecastings traffic related information (such as, as long as can obtain) can have precedence over forecast traffic related information, uses forecast traffic related information can have precedence over again history average traffic condition information.The additional detail relating to prediction and forecast future traffic flow situation can be submitted on March 3rd, 2006, and be entitled as the U.S. Patent application No.11/367 of " Dynamic Time Series Prediction Of Future Traffic Conditions ", obtain in 463, its full content is incorporated in this as a reference.In other embodiments, step 1215 and 1225 can not be performed, always perform based on the best data obtained from other healthy traffic sensor during selected time classification and/or relevant time classification if data readings such as in step 1220 corrects.Such as, if at least predetermined percentage of healthy traffic sensors all in the instruction road segment segment of selected time classification (such as, at least 30%) be healthy, then correct can based on these all traffic sensors for data readings, otherwise based on the healthy traffic sensor closed on most indicated by during select time classification and/or relevant time classification and/or in the road segment segment of closing on.

After step 1220 or 1225, routine proceeds to step 1230 and provides determined traffic sensor data readings as the correction reading for the traffic sensor during institute's classification seclected time.In certain embodiments, determined traffic sensor data readings can be the follow-up use of other assembly (such as, the sensor data collection assembly 110 of Fig. 1) and store (such as, being stored in database or file system).In step 1235, routine determines whether the additional combinations of traffic sensor and the time classification that will be processed.If had, then routine turns back to step 1210, if not, proceeds to step 1299 and terminates.

Figure 13 is the process flow diagram of the exemplary embodiment of sensor data readings gatherer routine 1300.This routine can be passed through, such as, perform the sensor data readings collection assembly 355 of Fig. 3 and/or Fig. 1 sensor data readings collection assembly 110 provides, such as, determine and be provided at special time classification or the traffic related information of multiple traffic sensor (the multiple traffic sensors such as associated with particular lane section) in section At All Other Times.In the illustrated exemplary embodiment, this routine is that particular lane section performs, but can collect information from multiple traffic sensor groups of other type in other embodiments.In addition, this routine can provide supplements by other routine of the estimation performing traffic related information (such as, data sample flow estimation device routine) traffic related information of information that provides, thus can not provide accurate estimation (such as due to data deficiencies) in other routine traffic related information is provided.

This routine starts in step 1305 and receives the instruction of one or more section and one or more time classification or section At All Other Times.In step 1310, routine selects next road segment segment of one or more indicated road segment segment from first.In step 1315, routine obtains some or all the available traffic sensor data readings gathered within the indicated time period by all traffic sensors associated with this road.Such information such as can obtain from the sensing data adjustment assembly 353 of the sensing data adjustment assembly 105 of Fig. 1 and/or Fig. 3.Particularly, routine can obtain traffic sensor data readings for being confirmed as healthy traffic sensor and/or from the traffic sensor data readings being confirmed as unsound traffic sensor and obtaining correction in some cases, and such as those are provided by the sensor data readings error corrector routine of Figure 12 or determine.

In step 1320, routine then collects obtained data readings in one or more ways, thus determines within the indicated time period for the average velocity of road segment segment, amount and/or occupancy.Average velocity can such as by determining through the data readings averaging of the car speed of one or more traffic sensor reflection.The volume of traffic can be determined according to the data readings of reporting vehicle quantity.Such as, given report is activated from sensor the loop sensor of the vehicle cumulative amount beginning through sensor, then the volume of traffic can by deducting two data readings obtaining within the indicated time period and removing result by the time interval between data readings and infer simply.In addition, density can be determined, as described in more detail elsewhere based on determined average velocity, amount and average Vehicle length.In some cases, data readings can weighting in every way (such as, passing through the age), so that data readings nearer in average discharge is determined has the impact larger than older data readings.

In step 1325, routine then determines whether that many road segment segment (or other group of multiple traffic sensor) will process.If had, then routine turns back to step 1310, otherwise proceeds to step 1330 to provide determined traffic flow information.In certain embodiments, determined flow information (such as, being stored in database or file system) can be stored to be follow-uply supplied to the traffic data clients 109 of Fig. 1 and/or the RT information providing system 363 of Fig. 3.Next, routine proceeds to step 1339 and returns.

Figure 14 is the process flow diagram of the exemplary embodiment of magnitude of traffic flow estimation device routine 1400.This routine can provide by such as performing magnitude of traffic flow estimation assembly (not shown), thus estimation is used for various types of traffic flow informations of road segment segment in every way.In the exemplified embodiment, such as when enough data can not be obtained for accurately performing their respective estimations when these routines, routine can by the data sample speed estimator routine call of Fig. 6 with obtains average velocity estimation and/or by the data sample flow estimation device routine call of Fig. 7 with the estimation of acquisition amount and/or occupancy.

This routine starts and reception channel section, one or more time classification or section At All Other Times in step 1405, and the instruction of the traffic flow information of one or more type such as one or more such as speed, amount, density, occupancy.In step 1410, routine determines whether the traffic flow information estimating instruction type based on one or more electronic map section, such as, whether there is within one or more shown time period based on such road segment segment the precise information of the traffic flow information for one or more type.Relevant road segment segment can identify in every way.Such as, in some cases, the information of electronic map section can comprise about the information of relation between road segment segment, such as first section usually has and is similar to second (such as, adjacent) the travel pattern of road segment segment, thus the traffic flow information for second section can be used for estimating the magnitude of traffic flow on first section.In some cases, no matter analysis is in advance and/or Dynamic Execution, can automatically determine such relation, such as based on magnitude of traffic flow pattern respective in the section of two road statistical study (such as, be similar to previously discussed about identifying that given traffic sensor is at the similar Data distribution8 of different time, but alternatively analyze in two or more different sensors in the similarity such as between the same time).Alternatively, one or more adjacent road segment segment can be selected to determine to associate indicated road segment segment without the need to any of particular kind of relationship between the road segment segment performed.If determine that based on related roads section estimation traffic flow information, then routine proceeds to step 1415 and estimates the value for the traffic flow information of indicated type based on the identical type traffic flow information for one or more related roads section.Such as, average traffic speed based on one or more neighboring track section determines the average velocity of this road segment segment (such as, by using the traffic speed come from a neighboring track section, or to the traffic speed averaging come from two or more neighboring track section).

If alternatively determine in step 1410 not based on the estimation of related roads section for the traffic flow information of indicated road segment segment, then routine proceed to step 1420 and determine whether within one or more indicated time period based on for this instruction road segment segment and instruction time section information of forecasting be indicated road segment segment estimation traffic flow information.In certain embodiments, such information of forecasting may only obtain on other occasions, if such as repeat prediction (such as ensuing 3 hours every 15 minutes once) for multiple future time instance to obtain accurate current data simultaneously.Similarly, if (such as, more than three hours) are available for generation of the accurate input data of prediction within time expand, then can without the need to obtaining the prediction of the future traffic condition information used by this routine.Alternatively, in certain embodiments, the future traffic condition information of such prediction is non-availability due to some other reason, such as, owing to not using in this embodiment.If determine based on information of forecasting estimation traffic flow information in step 1420, then routine proceeds to step 1425, and the information of forecasting providing system 360 to obtain based on the information of forecasting from such as Fig. 3 and be the instruction type of the road segment segment of instruction and the time period estimation traffic flow information of instruction.The additional detail relating to prediction and forecast future traffic flow situation is the U.S. Patent application No.11/367 being entitled as " Dynamic Time Series Prediction Of Traffic Conditions " submitted on March 3rd, 2006, can obtain in 463, its full content is incorporated in this as a reference.

If alternatively determine in step 1420 not based on information of forecasting be indicated road segment segment estimation traffic flow information (such as, because this information can not get), then routine proceeds to step 1430 and determines whether that based on the forecast information for this road segment segment and time period be indicated road segment segment estimation traffic flow information within the time period of one or more instruction.In certain embodiments, can the future time forecast traffic of predict traffic conditions for exceeding, such as, in the mode not using at least some current condition information.Similarly, if information of forecasting (such as, due to the accurate input data for generation of prediction more than three hours with regard to non-availability) can not be obtained, then still forecast information can be used, the information such as obviously produced in advance.If determine based on forecast information estimation traffic flow information in step 1430, then routine proceeds to step 1435 and is that indicated road segment segment and time period estimation indicate the traffic flow information of type based on the forecast information providing system 360 to obtain from such as predicted traffic information.

If be not alternatively that indicated road segment segment estimates traffic flow information (such as based on forecast information in step 1430, due to this information non-availability), then routine proceeds to step 1440 and is that indicated road segment segment and time period estimation indicate the traffic flow information of type (such as based on the history average discharge information for indicated road segment segment, for the identical or corresponding time period, such as, based on the time classification comprising moment at hour classification and/or date category).Such as, if forecast information be unavailable (such as, because the input data of the time longer than the cycle producing nearest prediction and forecast are unavailable, therefore can not produce new prediction and can not produce new forecast), then routine can use the history average discharge information for indicated road segment segment.Being entitled as in the U.S. Patent application (application attorney docket is 480234.410P1) of " Generating Repre sentative Road Traffic Flow Information From Historical Data " that the additional detail relating to generation history average discharge information can be submitted at the same time obtains, and its full content is incorporated in this as a reference.

After step 1415,1425,1435 or 1440, routine proceeds to step 1445 and provides the estimation traffic flow information of indicated type for indicated road segment segment and indicated time period.The information provided such as can be returned to the routine (such as, data sample flow estimation device routine) of calling this routine and/or be stored (such as, being stored in database or file system) in order to follow-up use.After step 1445, routine proceeds to step 1499 and returns.

Fig. 9 A-9C illustrates and is obtaining and providing the action example about the Mobile data source in road traffic condition information.The information of electronic map traffic can obtain from mobile device (no matter based on equipment or the subscriber equipment of vehicle) in every way, such as by using wireless link (such as, satellite uplink, cellular network, WI-FI, packet radio etc.) transmission and/or when equipment reaches suitable docking (docking) or other tie point physics carry out downloading (such as, once return operation main base or have can perform the suitable equipment that information is downloaded other object just from fleet's Download Info).Although provide various benefit (such as in the information of the electronic map traffic being obviously later than the very first time that the second time obtained, revise the prediction of the very first time, prediction processing etc. is improved) for using the data of institute's observed case subsequently, it can be such as the situation from equipment physics Download Info, when with in real time or be bordering on real-time mode obtain time, such road traffic condition information provides additional benefit.Therefore, in at least some embodiments, the mobile device with wireless communication ability can provide the information of the electronic map traffic needed at least some continually, such as periodically (such as, every 30 minutes, 1 minute, 5 minutes etc.) and/or when can the information needed of q.s available time (such as, for each data point relevant to road traffic condition information; For every N number of such data, such as wherein N is configurable number; Store when fetched data reaches specific and/or transmit size etc.).In certain embodiments, this radio communication frequently of the road traffic condition information obtained can also obtain road traffic condition information to supplement (such as by additional institute At All Other Times, download from the continuous physical of equipment, few frequency (less-frequency) radio communication via comprising greater amount data), such as comprise the additional data corresponding to each data point, comprise the collection information etc. about multiple data point.

Although by from mobile device with in real time or other frequently mode obtain obtained road traffic condition information and provide various benefit, such radio communication obtaining road traffic condition information can retrain in every way in certain embodiments.Such as, in some cases, from mobile device via specific radio link (such as, satellite is uploaded) transmission data cost structure can be with few frequency interval (such as, every 15 minutes) transmission that occurs, or mobile device can be previously programmed with such interval transmission.In some other situation, mobile device temporarily may lose the ability by transmission of radio links data, such as owing to lacking wireless coverage (such as in the region at mobile device place, cellular radio receiver base station owing to not closing on), due to other action performed by the user of mobile device or equipment, or due to the temporary transient problem of mobile device or associated transmitter.

Therefore, the mobile device that at least some is so in certain embodiments can be assigned or be configured to store multiple data sample (or so multiple data samples are stored in other associate device), can be transmitted together for use at least some information of multiple data sample in a wireless transmission.Such as, at least some mobile device is configured at mobile device by transmission of radio links data (not such as in certain embodiments, the each data sample of the usual individual transmission of mobile device, such as every 30 seconds or 1 minute) time the cycle memory storage road traffic condition information data sample that obtains, and then to transmit together at the time durations data sample that these are stored of the next wireless transmission of appearance.Some mobile devices can also be configured to performance period property (such as every 15 minutes, or when the data of specified amount can be used for transmission) wireless transmission, and obtain during the time interval between wireless transmission can also be configured at least some embodiments and the multiple data samples storing road traffic related information (such as with predetermined sampling rate, such as 30 seconds or one minute), and then the data sample that these store is transmitted together (or the subset of these samples and/or set) similarly during next wireless transmission.As an embodiment, if nearly the wireless transmission cost of 1000 information units is $ 0.25 and the size of each data sample is 50 units, then sampling per minute to send the data group (instead of per minute send each sample individually) comprising 20 samples for every 20 minutes be very helpful.In such embodiments, although data sample possibility slight delay is (in the example of cyclical transmission, be delayed the average half of the time period between transmission, assuming that regularly obtain data sample), then the road traffic condition information obtained from transmission still provides and is bordering on real-time information.And, can be produced based on multiple stored data sample by mobile device and additional information is provided in certain embodiments.Such as, if specific mobile device only can obtain the information about current present position during each data sample, but can not obtain the additional relevant information of such as speed and/or direction, then such additional relevant information can be calculated based on multiple follow-up data sample or be determined.

Particularly, Fig. 9 A describes has several interconnective road 925,930,935 and 940, with the exemplary area 955 (road 925 and 935 north-south is walked, and road 930 and 940 East and West direction is walked) of the legend instruction 950 in instruction road north orientation direction.Although only show the road of limited quantity, they can represent vast geographic area, such as, across several miles of interconnective expressways, or the subset of avenue across several district.In this example, Mobile data source (such as, vehicle, not shown) travel from position 945a to 945c within the cycle of 30 minutes, and be configured to obtain for every 15 minutes and transmit the data sample representing current traffic condition.Therefore, when Mobile data source starts to travel, it obtains at position 945a and transmits first data sample (as used in this example shown in asterisk " ★ "), obtain at position 945b after 15 minutes and transmit second data sample, and altogether obtaining at position 945c after 30 minutes and transmitting the 3rd data sample.In this example, each data sample comprises the instruction of current location (such as, in gps coordinate), current direction (such as, north orientation), present speed (such as, 30 minutes are per hour) and current time, represented by the transmission of the 945a of usage data value Pa, Da, Sa and Ta, and also can comprise out of Memory (such as, indicating the identifier in Mobile data source) alternatively.Whether the acquisition although it is so current traffic condition information provided provides more benefit, but can not determine multiple details from such data, comprise from the route of position 945b to 945c partly along road 930 or 940.And such sample data does not allow, such as using the part of the road 925 between 945a and 945b of position as reporting and the distinct road section of the different traffic predicted.

With with mode like Fig. 9 category-A, Fig. 9 B describes example 905, its Mobile data source within the cycle of 30 minutes runs over interconnective road 925,930,935 and 940 from position 945a to 945c, and Mobile data source sends the information (represented by the asterisk shown in position 945a, 945b and 945c) about traffic for every 15 minutes.But in this example, Mobile data source is configured to acquisition per minute and stores data sample, subsequent transmission is included in the first 15 minutes data from each data sample.Therefore, when Mobile data source travels between 945a and 945b of position, Mobile data source obtains the group 910b of 15 data sample 910b1 to 910b15, and in this example, utilizes the time of data sample to sentence the arrow of the direction in Mobile data source to indicate each data sample.In this example, each data sample comprises the instruction of current location, current direction, present speed and current time similarly, and comprises these each data values for data sample 910b in the continuous transmission of position 945b.Similarly, as Mobile data source travels between 945b and 945c of position, then Mobile data source obtains 15 data sample 910c1-910c15, and comprises each fetched data value for 15 data samples in the subsequent transmission of position 945c.By providing such additional data sample, various additional information can be obtained.Such as, be easy to now to determine that from the route of position 945b to 945c be partly along road 930 instead of road 940, and allow corresponding traffic related information to be used for road 930.In addition, the data sample that specific data sample is adjacent with them can provide the various information of electronic map smaller portions, the road 925 between 945a and 945b of position is such as allowed to be expressed as such as reaching 15 distinct road sections (such as, by each data sample is associated with distinct road section), it is each has the different road traffic condition of possibility.Such as, can observe out intuitively, average velocity for data sample 910b1-910b6 is roughly static (due to interval data sample substantially uniformly), and increase (because data sample corresponds to each position be gradually far apart out for the average velocity of data sample 9101-910b8, reflect the distance run between the data sample of this example of user in interval at given 1 minute and become large), and the average velocity of data sample 910b1-910b15 declines.Although data sample in this example directly provides the information about such speed, data message such in other embodiment can obtain from the data sample information only comprising current location.

Fig. 9 C describes the 3rd example 990, wherein Mobile data source runs over interconnective road sections from position 965a to 965c within the cycle of 30 minutes, and the information (as shown in asterisk in position 965a, 965b and 965c) about traffic is transmitted in Mobile data source for every 15 minutes.As shown in Fig. 9 C, Mobile data source is configured to acquisition per minute and stores data sample in this example, and subsequent transmission comprises each data from least some data sample in first 15 minutes.Therefore, as Mobile data source travels between 965a and 965b of position, then Mobile data source obtains the group 960b of 15 data sample 960b1-960b15.But, as the data sample 960b5-b13 by common location (owing to movement not detected for these data samples, therefore used annular instead of arrow in this example, but for the sake of clarity by its independent display instead of in top of one another), about 9 minutes (such as, stopping at cafe) has been stopped in Mobile data source in the side of road 925 in this embodiment.Therefore, when producing next transmission at position 965b, transmit all information that can comprise for all data samples in certain embodiments, or alternatively can omit at least some information (such as, the information of omitted data sample 960b6-960b12, if this is because know that Mobile data source is not still moved between data sample 960b5 and 960b13, then they do not provide additional useful information in this case).And, although do not illustrate here, but the information of one or more such data sample can be omitted in other embodiments, and follow-up transmission can be postponed until 15 data samples that will be transmitted are all available (such as, if based on the data volume that will be sent out instead of the property transmission of performance period time).And as Mobile data source travels between 965b and 965c of position, then Mobile data source obtains data sample 960c13 and 960c14 (as in this embodiment with opening shown in circle instead of arrow) in the current disabled region of radio communication.In other embodiments, wherein each data sample is individual transmission when acquisition but when not storing, and these data samples can be lost, but in this example, is store and transmit together with other data sample 960c1 to 960c12 on the contrary at position 965c.Although do not illustrate here, but in some cases Mobile data source can also temporarily lose usage data obtain basic device obtain the ability of one or more data sample (such as, if Mobile data source loses the ability a few minutes obtaining GPS reading) if---like this, then in certain embodiments Mobile data source can report other data sample obtained and without the need to further reaction (such as, if needed, allow take over party to insert or estimate these data samples), although can attempt in other embodiments otherwise to obtain data sample (such as, position is determined by using accurate not mechanism, such as cellular mobile telephone tower triangular is measured, or by based on previously known position and follow-up average velocity and orientation estimation current location, such as pass through dead reckoning), even if these data samples have lower accuracy or degree of accuracy (such as, can by comprising the degree of lower credibility to these data samples or higher possible errors, or how to produce by comprising instruction these and/or other data sample).

Although Fig. 9 B and 9C each in, example data sample illustrate only a vehicle or other Mobile data source for brevity, but in other embodiments, the multiple data samples for specific Mobile data source can not be used to determine the particular course gathered by this Mobile data source, and more specifically, even can not associate with each other (such as, if the source of each Mobile data sample is anonymous, or originate from other have nothing different).Such as, if the multiple data sample come from specific Mobile data source and can't help take over party be used in produce relate to these data samples collective data (such as, speed and/or directional information is produced) based on only providing the continuous print data sample of positional information, such as when such collection data comprise each data sample or not by use, such take over party can not be provided in certain embodiments to identify and to relate to Mobile data samples sources and/or indicate multiple data sample from identical Mobile data source (such as, determine to increase the privacy relating to Mobile data source based on design).

Alternatively, in the embodiment that at least some is such, multiple Mobile data source is used together to determine interested road condition information, such as, use the multiple data samples come from all Mobile data sources to determine the collection information of this road segment segment for particular lane section (or other parts of road).Like this, such as, the interested time period (such as, 1 minute, 5 minutes, 15 minutes etc.) in, the each of multiple incoherent Mobile data source can provide one or more data sample relating to its oneself traveling on particular lane section within this time period, and if each such data sample comprises speed and directional information (such as), then can determine average gathering speed for this time period and for the road segment segment of the movement in the same direction usually of all data sources, such as to be similar to the mode of the road traffic sensors for multiple vehicle collection information through sensor.Specific data sample can associate with particular lane section in every way, such as by data sample positions is associated (no matter for any road with the road (or road segment segment) with proximal most position, or only to the road meeting specific criteria, such as belong to the category of roads of one or more indicated function) and be then the road segment segment that this Model choices is suitable, or by using the instruction provided together with the data sample of associated road (or road segment segment) by Mobile data source.In addition, in at least some embodiments, in order to assign the object of data sample and other object (such as to road, using north orientation track, expressway as the different track different from the south orientation track of expressway), using the road except one-way road as distinct road, and if like this, the suitable road determining to associate with data sample then can also be used to for the direction of Mobile data sample---but, in other embodiments, can otherwise modeling, such as using two-way avenue as a road (such as, the average traffic situation reported according to the vehicle for movement in the two directions and predict), using each track of the expressway by multilane or other road as different logical road etc.

In certain embodiments, for the ease of using multiple Mobile data source to determine interested road condition information, fleet can be configured to provide used road sample in every way.Such as, if identical starting point is left in the similar time of every day by each large-scale fleet, then each portion vehicle can be configured differently into relate to and how soon and how long start to provide data sample, such as, minimize the change being in mass data all near single starting point and/or being provided in when obtaining and transmit data sample.More specifically, Mobile data source device can be configured to carry out how and when obtaining data sample in every way, comprise the total distance based on covering from starting point (such as the starting point of fleet's group), from the distance that last data sample obtains and/or covers transmission, (time that such as vehicle leaves from starting point) T.T. of experiencing from the outset, from the time that last data sample obtains and/or transmission is experienced, produce relevant one or more indicated position (such as, pass through, arrive, leave) indexical relation etc.Similarly, Mobile data source device can be configured to carry out how and when transmitting or provide one or more obtained data sample in every way, such as when adding up predetermined condition, comprise the total distance based on covering from starting point, the distance of covering from the acquisition of last data sample and/or transmission, the T.T. of experience is played from the outset, the time of experience from the acquisition of last data sample and/or transmission, produce the indexical relation about one or more indicated position, the instruction number of the multiple data samples collected, the indicated data volume of having collected (such as, fill up or fill up in fact the quantity of the buffer storing data sample on the mobile apparatus, or such as fill up or fill up in fact the quantity measured the instruction time for transmitting) etc.

Fig. 8 is the process flow diagram that Mobile data source information provides the exemplary embodiment of routine 800, such as can be provided for one or more data source 384 based on vehicle of Fig. 3 and/or each Mobile data source device based on the data source 101 of vehicle and/or other data source 102 of Fig. 1 of other data source 388 (such as, subscriber equipment) and/or Fig. 1 by operation.In this example, this routine is that specific Mobile data source obtains data sample to indicate current traffic, and suitably stores data sample so that subsequent transmission can comprise the information for multiple data source.

This routine starts in step 805, wherein retrieval will be used in the parameter of a part obtaining as data sample and provide, and such as configuration parameter is used to indicate the transmission that when should obtain data sample and when should produce the information corresponding to one or more data sample.Routine proceeds to step 810 and waits for, until obtain data sample in time, such as based on retrieved parameter and/or out of Memory (such as, through the time quantum indicated by past data sample acquisition, run over the shown distance of past data sample acquisition, indicated and obtain data sample etc. in continuous print mode in fact).Routine then proceeds to step 815 to obtain data sample based on current location and the mobile of Mobile data source, and stores data sample in step 820.If determine the time also not arriving transmission data in step 825, such as (such as measure through the instruction time of precedent transmission based on retrieved parameter and/or out of Memory, run over the instruction distance of precedent transmission, as long as indicate, it is available or transmit data sample etc. in continuous print mode in fact), then routine returns step 810.

Otherwise routine proceeds to step 830 to retrieve and to select any stored data sample due to precedent transmission (or from, from first time transmission).Routine then alternatively in step 835 based on multiple selected data sample (such as, for whole average velocitys of all data samples, if the information obtained only provides positional information, then for for the average velocity of each data sample and direction etc.) produce collected by data.But in other embodiments, also can not perform the generation of the data of such collection.In step 840, routine then alternatively from selected data sample group remove be used for some or all data samples some or all institute the information that obtain (such as, only transmission is used for the selected type of each data sample, remove the data sample that those occur exceptional value or mistake, remove those not corresponding to the data sample etc. of the actual movement in Mobile data source), in other embodiments, also can not perform such information to remove.In step 845, routine is then to the current information of take over party's transmission in current group of data sample and the information of any collection that will use by rights.In step 895, routine determines whether to continue (such as Mobile data source whether continue use and be moveable), and if be, then turns back to step 810.Otherwise routine proceeds to step 899 and terminates.In the embodiment can not transmitting data in Mobile data source and situation, no matter whether due to temporary transient situation or alternatively reflect the configuration first of Mobile data source, step 830-845 can not be performed until data can be transmitted in Mobile data source or provide (such as, via physics download) obtained due to previous transmission and some or all of the data sample stored.

Just as previously noted, once and obtain the information of electronic map traffic, such as from one or more Mobile data source and/or other source one or more, then can use road traffic condition information in every way, such as report present road traffic in substantially real-time mode, or use past and the next each prediction future traffic condition at multiple future time of current road traffic condition information.In certain embodiments, the future that the type of input data for generation of future traffic condition prediction can comprise various current, past and expect, and can comprise at predetermined time interval (such as from the output that prediction processing is come, three hours, or one day) in multiple future times each (such as, following every 5,15 or 60 minutes) interested multiple target track sections each on the prediction of expection traffic that produces, as described in more detail elsewhere.Such as, the type of input data can comprise following: about the information for the current of each target track section interested in geographic area and the past volume of traffic, such as, selected by the geographic area network of road; About the information of current and recent traffic hazard; About information that is current, recent and future trajectory engineering; About current, past with expect the information (such as, precipitation, temperature, wind direction, wind speed etc.) of external weather condition; About at least some is current, the information of past and the following event arranged (such as, the type of event, the start and end time of time expection, and/or the place of time or other position etc., such as all events, the event of instruction type, very great event, such as, have expection the attending of (such as 1000 or 5000 expection attendants) on indicated threshold value); With the information (such as, whether school attends class and/or the position of one or more school) about school schedules.In addition, although in certain embodiments, multiple future times of prediction future traffic condition are often points on time, but prediction so in other embodiments alternatively can represent multiple time point (such as, time period), such as, by representing the average or collection tolerance of the future traffic condition during these multiple time points.And, some or all of input data can be known and represent (such as by the degree determined of change, the weather of expection), and additional information can be produced represent in for produced prediction and/or the credibility of other metadata.In addition, being a variety of causes and can the prediction of initialization future traffic condition in each time, such as, in periodic manner (such as, every 5 minutes), when receiving any or enough new input data, responding from the request etc. for coming.

Some of identical type of input data can be used in certain embodiments to produce the forecast of longer-terms limit of future traffic condition similarly (such as, following one week, or following one month), but the forecast of such longer-term limit also can not use the input data of some types, such as about the information of the present situation (such as, Current traffic, weather or other situation) of the time in forecast generation.In addition, the forecast of such longer-term limit can produce with the forecast lower ground frequency comparing short-term limit, and can be produced and compare forecast that short-term limits and more can reflect different future time section (such as, per hour instead of every 15 minutes).

Can also select in every way for generation of future traffic condition prediction and/or forecast road and/or road segment segment.In certain embodiments, for multiple geographic area (such as, urban district) the prediction of each generation future traffic condition and/or forecast, wherein each geographic area has the road network of multiple interconnection---such geographic area can be selected in every way, can be such as a prominent question by (such as, based on the road traffic sensors network of at least some road in this region) and/or traffic congestion wherein easily based on current traffic condition information.In the embodiment that some are such, road for generation of future traffic condition prediction and/or forecast comprises the road that those are easy to obtain current traffic condition information, and in other embodiments, the selection of such road can at least in part based on one or more other factors (such as, based on size or the capacity of road, such as, comprise expressway and primary highway; Based on the road traffic regulation of carrying traffic, such as, comprise Class I highway and the blocked road that mainly can be substituted into the road of the such as larger capacity such as expressway and primary highway; Based on the functional category of road, such as, specified by federal expressway management board etc.).In other embodiments, can be that road produces future traffic condition prediction and/or forecast, and no matter its size and/or the mutual relationship with other road.In addition, can select in every way for generation of future traffic condition prediction and/or forecast road segment segment, such as using each road traffic sensors as different section; For each road segment segment and composition group that multiple road traffic sensors is put together (such as, reducing the quantity producing independent prediction and/or forecast, such as, by composition group of the road traffic sensors of specific quantity being put together); Select road segment segment to reflect the identical or logic relevant portion of road of fully similar (such as, strong association) of traffic; Such as based on from traffic sensor and/or other source (such as, from the data that vehicle and/or the user that travels at road produce, as elsewhere more in detail institute discuss) traffic related information; Deng.

In addition, future traffic condition can be used in various embodiments in a different manner to predict and/or forecast information, as elsewhere more in detail discuss, be included in each time in every way (such as, by by information transmission to cellular mobile telephone and/or other portable consumer device; By showing information to user, such as, by Web browser and/or application program; By information being supplied to other tissue and/or providing the entity of at least some information to user, such as analyzing and the third party etc. that provides of execution information after amendment information) such information is supplied to user and/or tissue (such as, response request, by periodicity transmission information etc.).Such as, in certain embodiments, usage forecastings and/or forecast information determine travel route and/or the time of suggestion, optimal time such as by travelling shown in the optimal route of road net and/or execution between starting position and final position, and such determination is based upon each prediction and/or the forecast information of multiple future times of one or more road and/or road segment segment.

In addition, various embodiment for user and other client provide various mechanism come with one or more traffic information system (such as, the data sample management system 350 of Fig. 3, RT information providing system 363, and/or information of forecasting provides system 360 etc.) mutual.Such as, some embodiments can ask and receive the corresponding client responded to provide interactive controlling (such as producing, client-side program provides mutual user interface, sing on web browser interface etc.), such as request relates to the information of current and/or predict traffic conditions and/or requirement analysis, selection, and/or provides the information relating to travel route.In addition, some embodiments provide API (" application programming interfaces "), and it allows client computing system programmably to carry out some or all requests, such as, by Network Message Protocol (such as, Web service) and/or other communication mechanism.

Those skilled in the art also can understand, and the function provided by routine as discussed previously in certain embodiments can provide in alternatively mode, such as, can be divided in multiple routine or focus on several routine.Similarly, shown in certain embodiments routine can provide than described more function, such as, when the routine shown in other alternatively lacks respectively or comprises such function, or when working as provided function selectable number.In addition, although various operation can (such as serial or parallel) and/or particular order perform as shown in a specific way, it will be understood by those skilled in the art in these operations in other embodiment and also can perform with other order and mode.Those skilled in the art is also accessible, and the data structure of above-mentioned discussion can build by different way, such as, concentrate in individual data segmentation of structures to multiple data structure or by multiple data structure in a data structure.Similarly, shown in certain embodiments data structure can store than described more or less information, such as, when the data structure shown in other alternatively lacks respectively or comprises such information, or when the amount or types of information that is stored is altered.

Being understandable that from above-mentioned, although there is described herein specific embodiment for the object of example, can various amendment being carried out when not deviating from the spirit and scope of the present invention.Therefore, the present invention is except claims and all not limited except this quotes element as proof.In addition, although particular aspects of the present invention is discussed with the form of given claim, inventor's imagination contains various aspects of the present invention with any available claim formats.Such as, although current only can being stated as in aspects more of the present invention is embedded in computer-readable medium, similar other side also can comprise.

Claims (30)

1. a method for computing machine execution, for estimating the data sample representing the vehicle travelled on road, described method comprises:
Receive the instruction of one or more road segment segment of one or more road, every bar road segment segment all has each multiple associated data samples reflecting the reported speeds of vehicle in described road segment segment; With
At least one each for described road segment segment:
Multiple associated data samples of this road segment segment of automatic analysis, to determine not represent the one or more of actual vehicle travel conditions in described road segment segment in those data samples, at least one of determined data sample is the statistics exceptional value relative to other data sample in multiple associated data sample; With
One or more instruction is provided to remove determined data sample from follow-up use, so that other data sample can be used for assisting the traveling in described road segment segment;
Wherein, one or more at least one road segment segment, determine that for road segment segment one or more data samples of the actual vehicle travel conditions do not represented in road segment segment comprise: at least in part based on the reported speeds that each data sample in data sample reflects, determine that described each data sample is the statistics exceptional value relative to other data sample in the multiple data samples associated with described road segment segment.
2. method according to claim 1, wherein, one or more at least one road segment segment, instruction is provided to comprise from the follow-up determined data sample of removal that uses: the associated data sample analyzing the road segment segment except determined data sample, to determine the average velocity of the vehicle travelled in road segment segment, indicate determined average velocity, to assist the traveling of other vehicle in road segment segment.
3. method according to claim 1, wherein, one or more at least one road segment segment, instruction is provided to comprise from the follow-up determined data sample of removal that uses: to analyze the road segment segment associated data sample except determined data sample, to determine the magnitude of traffic flow of the vehicle travelled in road segment segment, indicate the determined magnitude of traffic flow, to assist the traveling of other vehicle in road segment segment.
4. method according to claim 1, wherein, also performs following steps by removing an outlier detection: determine that the data sample associated with road segment segment is the statistics exceptional value relative to other data sample in the multiple data samples associated with described road segment segment.
5. method according to claim 1, wherein, determine that each of the one or more data samples associated with road segment segment also comprises relative to the statistics exceptional value of other data sample in the multiple data samples associated with described road segment segment:
The whole of multiple data samples for road segment segment determine average velocity and standard deviation; With
For one or more data samples each of road segment segment,
Based on for the determined average velocity of whole multiple data sample of road segment segment and standard deviation, determine average velocity and the standard deviation of other data samples all of road segment segment;
The reported speeds of determining data sample and for the difference in multiple data samples of road segment segment between the determined average velocity of other data samples all;
Definite threshold is carried out at least in part based on the standard deviation of other data samples all in multiple data samples of determined road segment segment; With
When determined difference exceeds determined threshold value, identify that this data sample is for statistics exceptional value.
6. method according to claim 1, wherein, each for described one or more road segment segment, also performs following steps in substantially real-time mode: determine one or more data samples of described road segment segment each be statistics exceptional value relative to other data sample in the multiple data samples associated with road segment segment.
7. method according to claim 1, wherein, one or more at least one road segment segment described, determine that one or more data samples of the road segment segment of the actual vehicle travel conditions that can not represent in road segment segment comprise: estimate that at least one determined data sample reflects the behavior of every portion vehicle of its reported speeds, and determine the behavior of estimated every portion vehicle and do not correspond to the actual vehicle travel conditions in road segment segment.
8. method according to claim 7, wherein, the estimation behavior of at least one vehicle corresponds to the vehicle stopped.
9. method according to claim 7, wherein, the behavior based at least one vehicle of one or more determined data sample corresponds to the travel conditions in road segment segment except the road segment segment associated with one or more determined data sample.
10. method according to claim 1, wherein, one or more at least one road segment segment, determine that one or more data samples of the road segment segment that can not represent actual vehicle travel conditions in road segment segment comprise: identify the multiple data samples reported by the single portion vehicle travelled in described road segment segment, determine the behavior of described single portion vehicle along with the time based on identified data sample, and determine that identified data sample does not represent the actual vehicle travel conditions in road segment segment based on determined behavior.
11. methods according to claim 1, wherein, one or more at least one road segment segment, determine that one or more data samples of the road segment segment of the actual vehicle travel conditions that can not represent in road segment segment comprise: the desired value identifying the multiple data samples associated with described road segment segment, and determine that determined data sample does not meet identified desired value.
12. methods according to claim 1, wherein, one or more at least one road segment segment, determine that one or more data samples of the road segment segment of the actual vehicle travel conditions that can not represent in road segment segment comprise: the statistical distribution determining the multiple data samples associated with described road segment segment, and determine that determined data sample does not meet determined statistical distribution.
13. methods according to claim 1, wherein, one or more at least one road segment segment, determine that one or more data samples of the road segment segment of the actual vehicle travel conditions that can not represent in road segment segment comprise: the multiple distinct data curves identifying described road segment segment, each data and curves reflects the different subset of the traveling state of vehicle gone up at least partially in described road segment segment, and determined data sample meets at least one data and curves of the subset reflecting uninterested traveling state of vehicle.
14. methods according to claim 13, wherein, at least one of the data and curves identified is Gaussian curve.
15. methods according to claim 1, wherein, one or more at least one road segment segment, the each of multiple associated data samples of described road segment segment also reflects the report time corresponding with the reported speeds of the vehicle of described data sample, the automatic analysis of multiple associated data samples of described road segment segment also corresponds to predetermined time section, so that actual vehicle travel conditions is travel conditions within a predetermined period of time in described road segment segment.
16. methods according to claim 15, wherein, one or more at least one road segment segment, determine that one or more data samples of the road segment segment of the actual vehicle travel conditions that can not represent in road segment segment comprise: identify that the report time of determined each data sample is not in the predetermined amount of time for described road segment segment.
17. methods according to claim 1, also comprise, each for multiple different time period, receive multiple associated data samples of one of described road segment segment, each associated data sample standard deviation reflects the reported speeds of the vehicle within the time period in this road segment segment described in report time place, based on the data sample of its report time within the described time period, the described time period each in be this road segment segment described perform automatic analysis.
18. methods according to claim 1, wherein, one or more at least one road segment segment, perform following steps in substantially real-time mode: the one or more data samples determining the road segment segment of the actual vehicle travel conditions that can not represent in road segment segment.
19. methods according to claim 18, wherein, the at least some of the multiple data samples associated with at least some of described road segment segment is obtained by the vehicle travelled in these road segment segment and is reported by described vehicle, obtain at least one data sample described one or more after, produce the report of described at least some data sample in substantially real-time mode.
20. 1 kinds of computing systems being configured to estimate the data sample representing driving vehicle, comprising:
First assembly, is configured to the every bar for multiple road, and receive the instruction of multiple data samples of described road, each data sample is reflected in the reported speeds of the vehicle that described road travels; With
Data sample exceptional value removes assembly, is configured to the every bar for multiple road:
Automatically determine multiple data samples of described road one or more are the statistics exceptional values of other data sample in multiple data samples relative to described road; With
Indicated data sample provides one or more instructions of multiple data samples of the described road except determined data sample, so that can assist the traveling on road;
Wherein, described first assembly and described data sample exceptional value removal assembly include for the software instruction for execution in the storer of described computing system.
21. computing systems according to claim 20, wherein, for at least one each of multiple road, determine that the data sample of road is that statistics exceptional value comprises relative to other data sample of the data sample of described road: perform removal outlier detection in substantially real-time mode, the instruction of the data sample of described road is provided to comprise: to analyze described data sample to determine the average velocity of the vehicle travelled on described road, and indicate determined average velocity, to assist the traveling of other vehicle on described road.
22. computing systems according to claim 20, wherein, described first assembly comprises receiving trap, for the every bar for multiple road, receive the instruction of multiple data samples of described road, each data sample reflects the reported speeds of the vehicle travelled on described road, described data sample exceptional value is removed assembly and is comprised device, for performing following operation: for every bar of multiple road, automatically determine multiple data samples of described road one or more are statistics exceptional values of other data sample of multiple data samples relative to described road, and provide except determined data sample, one or more instructions of multiple data samples of described road, so that indicated data sample can assist the traveling on road.
The method that 23. 1 kinds of computing machines perform, for estimating the data sample reported by the vehicle travelled on road, described data sample comprises the information of associated vehicle travel conditions, and described method comprises:
Receive the instruction of multiple road segment segment of one or more road;
Receive the information about the current traffic condition of described multiple road segment segment, the information received comprises multiple data sample, each data sample all reported from multiple vehicle, and reflects the reported speeds of a described vehicle at the reported position place of one of road segment segment described in report time place; With
Each for described multiple road segment segment, in the following manner, estimates the traffic of described road segment segment based on the data sample being identified the travel conditions representing described road segment segment:
The group of many data samples is identified, so that the data sample of this group has the reported position corresponding with the travel conditions of described road segment segment from multiple data sample;
For each data sample in this group, reported speeds based on these data samples determines average velocity and the standard deviation of other data samples all of this group, and based on the difference between the reported speeds of data sample and determined average velocity and determined standard deviation difference how, determine whether described data sample is statistics exceptional value for other data sample of this group;
The data sample being confirmed as adding up exceptional value is removed from this group; With
After the removal, remaining data sample in this group is used to be that traffic inferred by all vehicles travelled in road segment segment,
Can be used for assisting the traveling in described road segment segment to make the traffic inferred based on data sample.
24. methods according to claim 23, wherein, multiple different time period each in, for the estimation of each execution traffic of multiple road segment segment, and wherein also perform the identification of the group of the data sample of road segment segment within the time period, so that the data sample identified of this group has the report time corresponding with this time period.
25. methods according to claim 23, wherein, for at least one each of multiple road segment segment, being confirmed as a data sample of the identification group of the road segment segment being statistics exceptional value, is from the vehicle that another road segment segment travels and the data sample associated with a described road segment segment improperly.
26. methods according to claim 23, wherein, for at least one each of multiple road segment segment, being confirmed as a data sample of the identification group of the road segment segment being statistics exceptional value, is from being parked in described road segment segment or the data sample of the vehicle on described road segment segment side.
27. methods according to claim 23, wherein, for each estimation traffic of described multiple road segment segment comprises: the average velocity and the standard deviation that determine all data samples of institute's identification group of described road segment segment, use and be used for the determined average velocity of all data samples and standard deviation, as each data sample of identification group determine the average velocity of other data samples all and the part for standard deviation of this group.
28. methods according to claim 27, wherein, determine that whether the data sample organized is that statistics exceptional value comprises relative to other data sample of this group: based on the determined standard deviation for this other data sample of group, determine that whether difference between the reported speeds and the average velocity of this other data sample of group determined of data sample is beyond threshold value.
29. methods according to claim 23, wherein, repeat to receive the data sample relevant with the current traffic condition of described multiple road segment segment, to reflect the change of traffic, perform the estimation to each traffic of described multiple road segment segment for the data sample recently received in real time fashion.
30. methods according to claim 29, wherein, in use group, remaining data sample infers that the traffic of all vehicles travelled in road segment segment comprises: for remaining data sample determination average velocity, infer average velocity based on determined average velocity for all vehicles travelled in described road segment segment, and the information of the average velocity about inferring is supplied to the one or more people considering to travel in described road segment segment.
CN201110221617.2A 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources CN102394008B (en)

Priority Applications (19)

Application Number Priority Date Filing Date Title
US77894606P true 2006-03-03 2006-03-03
US60/778,946 2006-03-03
US78974106P true 2006-04-05 2006-04-05
US60/789,741 2006-04-05
US11/431,980 2006-05-11
US11/432,603 2006-05-11
US11/432,603 US20070208501A1 (en) 2006-03-03 2006-05-11 Assessing road traffic speed using data obtained from mobile data sources
US11/431,980 US20070208493A1 (en) 2006-03-03 2006-05-11 Identifying unrepresentative road traffic condition data obtained from mobile data sources
US11/438,822 US7831380B2 (en) 2006-03-03 2006-05-22 Assessing road traffic flow conditions using data obtained from mobile data sources
US11/438,822 2006-05-22
US11/444,998 US8014936B2 (en) 2006-03-03 2006-05-31 Filtering road traffic condition data obtained from mobile data sources
US11/444,998 2006-05-31
US11/473,861 US7912627B2 (en) 2006-03-03 2006-06-22 Obtaining road traffic condition data from mobile data sources
US11/473,861 2006-06-22
US83870006P true 2006-08-18 2006-08-18
US60/838,700 2006-08-18
US11/540,342 US7706965B2 (en) 2006-08-18 2006-09-28 Rectifying erroneous road traffic sensor data
US11/540,342 2006-09-28
CN200780015916.22007.03.02 2007-03-02

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN 200780015916 Division CN101438335B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources

Publications (2)

Publication Number Publication Date
CN102394008A CN102394008A (en) 2012-03-28
CN102394008B true CN102394008B (en) 2015-01-07

Family

ID=38181159

Family Applications (6)

Application Number Title Priority Date Filing Date
CN201110221617.2A CN102394008B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN 201110220737 CN102254434B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN201110221624.2A CN102394009B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN 200780015916 CN101438335B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN201110221478.3A CN102289935B (en) 2006-03-03 2007-03-02 Use the data estimation road traffic condition from Mobile data source
CN201110221620.4A CN102289936B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources

Family Applications After (5)

Application Number Title Priority Date Filing Date
CN 201110220737 CN102254434B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN201110221624.2A CN102394009B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN 200780015916 CN101438335B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources
CN201110221478.3A CN102289935B (en) 2006-03-03 2007-03-02 Use the data estimation road traffic condition from Mobile data source
CN201110221620.4A CN102289936B (en) 2006-03-03 2007-03-02 Assessing road traffic conditions using data from mobile data sources

Country Status (7)

Country Link
EP (2) EP2278573B1 (en)
JP (1) JP2009529187A (en)
CN (6) CN102394008B (en)
AT (1) AT523869T (en)
AU (1) AU2007224206A1 (en)
ES (2) ES2373336T3 (en)
WO (1) WO2007103180A2 (en)

Families Citing this family (83)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6587781B2 (en) 2000-08-28 2003-07-01 Estimotion, Inc. Method and system for modeling and processing vehicular traffic data and information and applying thereof
US7620402B2 (en) 2004-07-09 2009-11-17 Itis Uk Limited System and method for geographically locating a mobile device
JP4924407B2 (en) 2007-12-25 2012-04-25 富士通株式会社 Sensor diagnostic method and sensor diagnostic apparatus
JP4935704B2 (en) * 2008-02-14 2012-05-23 アイシン・エィ・ダブリュ株式会社 Parking lot congestion state determination device, parking lot congestion state determination method, and computer program
JP4983660B2 (en) 2008-03-14 2012-07-25 アイシン・エィ・ダブリュ株式会社 Navigation system and route search method
GB0901588D0 (en) 2009-02-02 2009-03-11 Itis Holdings Plc Apparatus and methods for providing journey information
JP5378002B2 (en) * 2009-02-19 2013-12-25 アイシン・エィ・ダブリュ株式会社 Vehicle motion estimation device, vehicle motion estimation method, and vehicle motion estimation program
DE112009004441T5 (en) * 2009-03-03 2012-08-09 Toyota Jidosha K.K. The vehicle drive assist device
CN101694743B (en) 2009-08-25 2011-09-21 北京世纪高通科技有限公司 Method and device for predicting road conditions
CN102032911B (en) * 2009-09-29 2014-05-28 宏达国际电子股份有限公司 Vehicle navigation method, system and computer program product
CN102262819B (en) 2009-10-30 2014-10-15 国际商业机器公司 Method and device for determining real-time passing time of road based on mobile communication network
US9406229B2 (en) * 2009-11-12 2016-08-02 Gm Global Technology Operations, Llc Travel lane advisor
US8620569B2 (en) 2009-12-17 2013-12-31 Toyota Jidosha Kabushiki Kaisha Vehicle control device
CN101782396B (en) * 2010-03-05 2011-12-28 中国软件与技术服务股份有限公司 A navigation method and a navigation system
JPWO2011125185A1 (en) * 2010-04-07 2013-07-08 トヨタ自動車株式会社 Vehicle travel support device
JP5195848B2 (en) * 2010-08-31 2013-05-15 株式会社デンソー Traffic situation prediction device
KR101506927B1 (en) * 2010-09-16 2015-04-06 에스케이플래닛 주식회사 System for collecting of traffic information, revision device of valid sampling and method for measurement of each average velocity of group, and recording medium thereof
KR101776807B1 (en) 2010-09-16 2017-09-19 에스케이텔레콤 주식회사 System for collecting of traffic information, revision device of valid sampling and method for measurement of velocity, and recording medium thereof
CN102446413A (en) * 2010-09-30 2012-05-09 西门子公司 Method and device for acquiring path information based on switching information of mobile terminal
BE1019524A3 (en) * 2010-09-30 2012-08-07 Be Mobile Nv System and method for travel time measurement.
KR101508136B1 (en) * 2010-10-22 2015-04-06 에스케이플래닛 주식회사 System for collecting of traffic information, revision device of valid sampling and method for revision of valid sampling
JP5601177B2 (en) * 2010-11-30 2014-10-08 アイシン・エィ・ダブリュ株式会社 Position specifying device, position specifying method, and position specifying program
CN102044153B (en) * 2010-12-22 2015-11-25 广州通易科技有限公司 The method and system of the unobstructed grade evaluation of a kind of traffic flow
JP5739182B2 (en) * 2011-02-04 2015-06-24 インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation Control system, method and program
JP5731223B2 (en) 2011-02-14 2015-06-10 インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation Abnormality detection device, monitoring control system, abnormality detection method, program, and recording medium
JP5689333B2 (en) 2011-02-15 2015-03-25 インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation Abnormality detection system, abnormality detection device, abnormality detection method, program, and recording medium
CN102157070B (en) * 2011-03-31 2013-01-09 天津大学 Road traffic flow prediction method based on cell phone data
CN102169630B (en) * 2011-03-31 2013-04-24 上海电科智能系统股份有限公司 Quality control method of road continuous traffic flow data
GB2492369B (en) 2011-06-29 2014-04-02 Itis Holdings Plc Method and system for collecting traffic data
CN102360529B (en) * 2011-09-20 2013-07-24 张忠义 Method for directly acquiring traffic speed of urban road
CN102521985A (en) * 2012-01-06 2012-06-27 北京捷易联科技有限公司 Traffic information evaluation method and device
WO2013108406A1 (en) * 2012-01-20 2013-07-25 トヨタ自動車 株式会社 Vehicle behavior prediction device and vehicle behavior prediction method, and driving assistance device
CN102568207A (en) * 2012-02-02 2012-07-11 北京捷易联科技有限公司 Traffic data processing method and device
US9132742B2 (en) * 2012-02-23 2015-09-15 International Business Machines Corporation Electric vehicle (EV) charging infrastructure with charging stations optimumally sited
GB201205348D0 (en) 2012-03-27 2012-05-09 Tomtom Belgium Nv Digital location-based data methods and product
CN102779410B (en) * 2012-07-19 2014-08-06 杭州师范大学 Parallel implementation method of multi-source heterogeneous traffic data fusion
CN102867421B (en) * 2012-09-24 2014-07-09 东南大学 Method for identifying outlier data in effective parking lot occupancy
US20140143184A1 (en) * 2012-11-21 2014-05-22 Microsoft Corporation Turn restriction inferencing
CN103888893A (en) * 2012-12-19 2014-06-25 中国移动通信集团公司 System and method for obtaining traffic state information
US9195938B1 (en) 2012-12-27 2015-11-24 Google Inc. Methods and systems for determining when to launch vehicles into a fleet of autonomous vehicles
CN104952270A (en) * 2012-12-28 2015-09-30 观致汽车有限公司 Traffic jam message advance notice method and system
US9014957B2 (en) 2012-12-29 2015-04-21 Google Inc. Methods and systems for determining fleet trajectories to satisfy a sequence of coverage requirements
EP2948932A1 (en) 2013-01-24 2015-12-02 Roger Andre Eilertsen A traffic surveillance and guidance system
CN104123833B (en) * 2013-04-25 2017-07-28 北京搜狗信息服务有限公司 A kind of planning method and device of condition of road surface
KR101338496B1 (en) * 2013-07-18 2013-12-10 주식회사 로드코리아 Load monitoring method
CN104376712B (en) * 2013-08-16 2017-10-13 株式会社日立制作所 Lack transport information complementing device and its method
CN103473928B (en) * 2013-09-24 2015-09-16 重庆大学 Based on the urban traffic blocking method of discrimination of RFID technique
CN104580292B (en) * 2013-10-16 2019-03-05 电信科学技术研究院 A kind of acquisition methods of running condition information, report method and equipment
MX344376B (en) * 2013-10-17 2016-12-13 Ford Global Tech Llc Road characteristic prediction.
CN103561123B (en) * 2013-10-28 2017-05-10 北京国双科技有限公司 Method and device for determining IP segment affiliation
JP5613815B1 (en) * 2013-10-29 2014-10-29 パナソニック株式会社 Residence status analysis apparatus, residence status analysis system, and residence status analysis method
CN103578274B (en) * 2013-11-15 2015-11-04 北京四通智能交通系统集成有限公司 A kind of traffic flow forecasting method and device
EP2887332B1 (en) * 2013-12-23 2016-09-07 Siemens Aktiengesellschaft Method and system for detection of a traffic situation on a stretch of road
CN105091889B (en) * 2014-04-23 2018-10-02 华为技术有限公司 A kind of determination method and apparatus of hotspot path
CN105303820A (en) * 2014-07-18 2016-02-03 中兴通讯股份有限公司 Traffic condition information providing method and device and server
CN104182618B (en) * 2014-08-06 2017-06-30 西安电子科技大学 A kind of method for early warning that knocks into the back based on Bayesian network
JP6079721B2 (en) 2014-08-07 2017-02-15 トヨタ自動車株式会社 Vehicle driving support system
EP3186662B1 (en) * 2014-08-26 2019-03-20 Microsoft Technology Licensing, LLC Measuring traffic speed in a road network
KR101673307B1 (en) 2014-12-19 2016-11-22 현대자동차주식회사 Navigation system and path prediction method thereby, and computer readable medium for performing the same
JP6229981B2 (en) 2014-12-26 2017-11-15 パナソニックIpマネジメント株式会社 Vehicle detector abnormality detection device, traffic condition analysis device, vehicle detector abnormality detection system, traffic condition analysis system, and program
CN104599499B (en) * 2015-01-12 2017-08-29 北京中交兴路车联网科技有限公司 A kind of method and device of distributed statistics traffic location
CN104778834B (en) * 2015-01-23 2017-02-22 哈尔滨工业大学 Urban road traffic jam judging method based on vehicle GPS data
CN106156966A (en) 2015-04-03 2016-11-23 阿里巴巴集团控股有限公司 Logistics monitoring method and equipment
CN104751644B (en) * 2015-04-14 2017-02-22 无锡物联网产业研究院 Traffic detection method and traffic detection device
CN104933860B (en) * 2015-05-20 2017-07-11 重庆大学 Bus traffic congestion delay time at stop Forecasting Methodology based on gps data
US9911327B2 (en) 2015-06-30 2018-03-06 Here Global B.V. Method and apparatus for identifying a split lane traffic location
US9640071B2 (en) 2015-06-30 2017-05-02 Here Global B.V. Method and apparatus for identifying a bi-modality condition upstream of diverging road segments
CN108140320A (en) * 2015-10-30 2018-06-08 三菱电机株式会社 Notify control device and notification control method
CN105404772A (en) * 2015-11-04 2016-03-16 成都天衡电科科技有限公司 Segmented system memory effect based adaptive historical data analysis method
CN105303833B (en) * 2015-11-05 2017-06-20 安徽四创电子股份有限公司 Overpass accident method of discrimination based on microwave vehicle detector
CN105389980B (en) * 2015-11-09 2018-01-19 上海交通大学 Short-time Traffic Flow Forecasting Methods based on long short-term memory recurrent neural network
CN106683447B (en) * 2015-11-11 2019-11-19 中国移动通信集团公司 A kind of traffic lamp control method and device
CN105575155B (en) * 2016-01-08 2018-09-18 上海雷腾软件股份有限公司 Method and apparatus for determining vehicle traveling information
CN105486322A (en) * 2016-01-14 2016-04-13 上海博泰悦臻网络技术服务有限公司 Regional road condition information acquisition method and system
CN105679055A (en) * 2016-04-01 2016-06-15 深圳市智汇十方科技有限公司 Road status prediction method and prediction system
US10068470B2 (en) 2016-05-06 2018-09-04 Here Global B.V. Determination of an average traffic speed
US10147315B2 (en) 2016-07-27 2018-12-04 Here Global B.V. Method and apparatus for determining split lane traffic conditions utilizing both multimedia data and probe data
US10198941B2 (en) 2016-07-27 2019-02-05 Here Global B.V. Method and apparatus for evaluating traffic approaching a junction at a lane level
CN106781489B (en) * 2016-12-29 2019-07-26 北京航空航天大学 A kind of road network trend prediction method based on recurrent neural network
US10262479B2 (en) * 2017-02-24 2019-04-16 Huf North America Automotive Parts Mfg. Corp. System and method for communicating with a vehicle
CN106846816B (en) * 2017-04-12 2019-09-17 山东理工大学 A kind of discretization traffic state judging method based on deep learning
CN107180530B (en) * 2017-05-22 2019-09-06 北京航空航天大学 A kind of road network trend prediction method based on depth space-time convolution loop network
CN107978153B (en) * 2017-11-29 2019-07-26 北京航空航天大学 A kind of multimode traffic factors influencing demand analysis method based on space vector autoregression model

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1645402A (en) * 2005-01-19 2005-07-27 北京交通大学 Road traffic flow data quality controlling method and apparatus

Family Cites Families (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5590217A (en) * 1991-04-08 1996-12-31 Matsushita Electric Industrial Co., Ltd. Vehicle activity measuring apparatus
DE19631630C2 (en) 1996-08-05 2001-05-31 Homag Maschinenbau Ag Processing machine with a Saugspanneinrichtung
JPH10160494A (en) * 1996-11-29 1998-06-19 Toyota Motor Corp On-vehicle communication processing system
AU7703598A (en) * 1997-05-30 1998-12-30 David S. Booth Generation and delivery of travel-related, location-sensitive information
DE19805869A1 (en) * 1998-02-13 1999-08-26 Daimler Chrysler Ag Method and apparatus for determining the traffic situation on a road network paths
CA2290301A1 (en) 1999-03-05 2000-09-05 Loran Network Management Ltd. A method for detecting outlier measures of activity
CA2266208C (en) * 1999-03-19 2008-07-08 Wenking Corp. Remote road traffic data exchange and intelligent vehicle highway system
DE19928082C2 (en) * 1999-06-11 2001-11-29 Ddg Ges Fuer Verkehrsdaten Mbh Filtering methods or timetables for determining travel speeds and domain residual velocities
US6490519B1 (en) * 1999-09-27 2002-12-03 Decell, Inc. Traffic monitoring system and methods for traffic monitoring and route guidance useful therewith
US6989765B2 (en) * 2002-03-05 2006-01-24 Triangle Software Llc Personalized traveler information dissemination system
JP3990641B2 (en) * 2002-03-27 2007-10-17 松下電器産業株式会社 Road information providing system and apparatus and road information generation method
US20040034467A1 (en) * 2002-08-09 2004-02-19 Paul Sampedro System and method for determining and employing road network traffic status
WO2004021305A2 (en) * 2002-08-29 2004-03-11 Itis Holdings Plc Apparatus and method for providing traffic information
JP3874745B2 (en) * 2003-01-22 2007-01-31 松下電器産業株式会社 Traffic information providing method, traffic information providing system and apparatus
JP2004280521A (en) * 2003-03-17 2004-10-07 Matsushita Electric Ind Co Ltd Method and device for transmitting traveling track in probe car system
JP4255007B2 (en) * 2003-04-11 2009-04-15 株式会社ザナヴィ・インフォマティクス Navigation device and travel time calculation method thereof
JP4392747B2 (en) 2003-12-24 2010-01-06 アイシン・エィ・ダブリュ株式会社 Navigation system
CN100337256C (en) * 2005-05-26 2007-09-12 上海交通大学 Method for estimating city road network traffic flow state

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1645402A (en) * 2005-01-19 2005-07-27 北京交通大学 Road traffic flow data quality controlling method and apparatus

Also Published As

Publication number Publication date
EP1938296B1 (en) 2011-09-07
WO2007103180A3 (en) 2007-12-06
WO2007103180A2 (en) 2007-09-13
CN102289935A (en) 2011-12-21
EP1938296A2 (en) 2008-07-02
CN102394009A (en) 2012-03-28
CN101438335A (en) 2009-05-20
AU2007224206A1 (en) 2007-09-13
CN102254434B (en) 2013-05-01
ES2373336T3 (en) 2012-02-02
CN102289935B (en) 2015-12-16
CN102289936A (en) 2011-12-21
CN102254434A (en) 2011-11-23
AT523869T (en) 2011-09-15
EP2278573A1 (en) 2011-01-26
CN102394008A (en) 2012-03-28
CN102289936B (en) 2014-08-06
JP2009529187A (en) 2009-08-13
ES2386529T3 (en) 2012-08-22
CN102394009B (en) 2014-05-14
EP2278573B1 (en) 2012-05-16
CN101438335B (en) 2011-09-21

Similar Documents

Publication Publication Date Title
Zhan et al. Urban link travel time estimation using large-scale taxi data with partial information
US7706964B2 (en) Inferring road speeds for context-sensitive routing
Aslam et al. City-scale traffic estimation from a roving sensor network
CA2777931C (en) System for monitoring vehicle and operator behavior
US8358222B2 (en) GPS-generated traffic information
US9958280B2 (en) Assessing inter-modal passenger travel options
US8615354B2 (en) Displaying road traffic condition information and user controls
US7499949B2 (en) Method and system for obtaining recurring delay data using navigation systems
Yuan et al. T-drive: driving directions based on taxi trajectories
US6574548B2 (en) System for providing traffic information
Qureshi et al. A survey on intelligent transportation systems
CA2293536C (en) Method and device for generating, merging and updating of destination tracking data
CN101297337B (en) Methods for predicting destinations from partial trajectories employing open-and closed-world modeling methods
US8041660B2 (en) Optimizing traffic predictions and enhancing notifications
US7885758B2 (en) GPS-based traffic monitoring system
CN102939623B (en) Based on driver's action learning road navigation track of set
US20120173530A1 (en) Method for Differentiating Traffic Data Obtained from Probe Vehicles
Yoon et al. Surface street traffic estimation
CN101964148B (en) Road traffic information recording server and GPS (Global Positioning System) user terminal
US20100211304A1 (en) Personalized User Routing and Recommendations
JP4733251B2 (en) Data collection method and system using mobile phone location matched with road network
US9324232B2 (en) Method and system for modeling and processing vehicular traffic data and information and applying thereof
US20030014181A1 (en) Traffic information gathering via cellular phone networks for intelligent transportation systems
US9816823B2 (en) Updating road maps
JP5291935B2 (en) Apparatus and method for predicting future movement of an object

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20150107

Termination date: 20180302