EP1973086B1 - Method and system for recognising traffic disturbances in a dynamically changing traffic environment - Google Patents

Abstract

The method involves detecting signals from a temporal sequence by a sensor that is attached to a vehicle, where the signals indicate the presence of the vehicle in a spatial area. A traffic route is inferred by the evaluation of the temporal sequence of the signals on the uninterrupted locomotion of the vehicle. Occurrence of traffic disturbances is detected and indicated, which is complementary to the evaluation of the temporal sequence of the signals based on criteria. An independent claim is also included for a system for detecting traffic disturbances in a dynamically changing traffic situation in a spatial area.

Description

The invention relates to a method for detecting traffic congestion in a dynamically changing traffic situation according to the preamble of claim 1 and a system performing the method according to the preamble of the independent device claim.

The traffic situation on our roads is very much influenced by private traffic and can occur, for example. during a day with a high dynamic change, so that unpleasant traffic disruptions, such as inhibiting the flow of traffic and especially congestion can occur very quickly. The early recognition of such traffic disruptions and even more their premature prognosis place high technical demands on systems and services designed for them.

Among the conventional traffic reporting services, the collective information services that are processed via the broadcasting stations or broadcasters are known, which inform the drivers concerned about traffic jams and other traffic disruptions in their respective broadcasting area. Also, increasingly, the information is digitally transmitted e.g. transmitted via mobile to installed in the vehicles navigation systems, which not only provide the respective motorists with the current traffic information, but also offer him suggestions for bypassing such traffic disruptions.

These and similar traffic information can be distributed via various media (telephone, Internet, data services), whereby they are centrally recorded or processed and usually consist of individual traffic reports, which in turn by direct observation of the objects of This is mainly due to human observers, such as police officers and so-called traffic jam detectors, who detect a traffic jam and then report it by radio or telephone to a central office. There are also technical devices in the form of stationary sensors (infrared sensors, ultrasonic sensors, induction loops, ...) or mobile sensors, which are installed in so-called "floating cars". The sensors detect low speeds or high traffic densities and report them to the control center. In the center then the respective position, length and possibly other characteristics of the traffic disturbance is determined from the sum of the individual observations and their time course, possibly even from a single observation, expressed as a traffic message and provided for individual information services or as collective traffic information , In these and other known methods, for example, in the article " Stau-Schau "by P.-M. Ziegler and U. Fastenrath in the magazine" ct "issue 9/2004 on pages 172 - 181 Referenced.

Although numerous methods for the detection or collection of traffic data are known, sometimes the completeness of the information thus obtained leaves something to be desired. This is especially true for those parts of the road network which are prone to traffic jams but where there is no stationary detection such as e.g. on federal roads. This also applies to roads where the average traffic flow and thus the density of mobile sensors is lower than on federal motorways. There is currently no suitable source of quality and complete traffic information for these subnets.

In the DE 19638798 A1 a system-based method for traffic data acquisition via mobile networks is proposed. In this case, an evaluation of the application-specific, coarse information about the respective is done to generate traffic information Whereabouts of mobile phones as they exist in any mobile network. Although this proposal seems to solve the problem of completeness - since a sufficient number of mobile phones on all road classes in vehicles moves - but it brings with it special problems: So, in contrast to conventional, application-specific sources of traffic data first of all to clarify whether the Sensor, namely the mobile phone, even delivers valid data on individual transport or possibly in a bus, by bike, on foot or not at all. Especially in the interesting situations, ie at low flow velocities in a traffic jam, this distinction is naturally enormously difficult. In addition, the location information present in the mobile network is typically associated with errors of several hundred meters, so that the exact assignment of measured speeds to the road network, represented by a digital map, especially in dense networks, especially in urban networks, difficult and often impossible is. Thus, even with this data source a clear signal for disturbed traffic can take a long time to wait, possibly failing despite an existing traffic incident. In particular, it remains to solve the problem of extracting from the existing data sources complete, up-to-date information on location, length and severity of traffic disruptions that occur in subnets that are not monitored by stationary sensors.

From the US 6,810,321 A method is known in which the vehicles communicate their speed at regular intervals, from which then an average speed is calculated. At each time point, an average speed at that location is calculated from a plurality of speeds at a location. The successive average speeds calculated in this way are deducted from each other, and when the threshold falls below a predetermined threshold, a congestion is assumed.

From the EP 1 732 188 A2 For example, a method is known in which a defined vehicle with a GPS receiver periodically transmits information to a center, distinguishing whether the driver speaks or does not speak. During speech, there is a connection so that the data of the location of the vehicle can be transmitted, whereby the data are obtained by pollen from the mobile device during non-speech.

From the EP 1 235 195 A2 For example, a method is known in which a specific terminal within the vehicle collects, as a floating car, information that is sent to a central office. Thus, only specific vehicles with corresponding terminals are to be considered.

The Where 20041059592 also works on the basis of a sample vehicle, which must send specific information to a central office. It thus requires a specific terminal in a vehicle.

It is therefore an object of the present invention to provide a method and a system for detecting traffic congestion in a dynamically changing traffic situation, which overcome the aforementioned disadvantages of the conventional solutions in an advantageous manner. This object is achieved by a device and a method according to the independent claims.

In particular, in order to achieve the above object, the present invention provides a method for detecting traffic congestion in a dynamically changing traffic situation in a spatial area in which at least one of a vehicle associated mobile device signals are detected in chronological order, the presence of the View vehicle in the spatial area, wherein by an evaluation of the temporal sequence of the signals on a continuous movement of the vehicle along a located in the spatial area traffic route can be closed or is, and in which complementary to the evaluation of the temporal sequence of the signals based on the absence of the signals, an occurrence of traffic disturbances is detected and displayed. Here, the typical frequencies of the signals for free traffic depending on route section, time of day and day of the week are determined from historical data, which is determined from the frequency response a minimum time that must be waited before a traffic incident is logged.

Accordingly, there is an easy to implement, yet very reliable method of detecting traffic congestion. Because is essentially on existing data or signals from mobile sensors, in particular mobile phones, resorted, from which with high probability on the state "free ride" can be concluded, with an intelligent evaluation of the traffic disturbances recognized as space-time complement and can be displayed.

In addition, the present invention provides a system implementing the method, the system comprising receiving devices and a center connected thereto, which receives, at least from a mobile sensor associated with a vehicle, signals in chronological order which indicate a presence of the vehicle in the vehicle Show spatial area, which can be concluded by an evaluation of the temporal sequence of the signals on a continuous movement of the vehicle along a located in the spatial area traffic route, and that the central complementary to the evaluation of the temporal sequence of signals based on at least one criterion Identify traffic incidents and prepared for display or for forwarding to processing systems or users.

Advantageous embodiments of the invention can be found in the subclaims:

Accordingly, it is advantageous if the at least one criterion is the absence of the signals to be detected during a predefinable period of time. For the at least temporary absence of the signals indicates a no longer present or static presence of the mobile sensors (and thus the vehicles) in the observed space, which in turn indicates that there is a high probability of no free ride for these vehicles and thus here a traffic incident is indicated.

Alternatively or in addition to this, the at least one criterion may be the time profile of the sequence of the signals to be detected. In this context, it is advantageous if the at least one criterion represents a sign and / or logoff of the vehicle in the spatial area signaled by the respective mobile sensor. Here, it is advantageous to focus on typical signals which are usually signaled in mobile cellular networks in connection with handover procedures.

Accordingly, it is also advantageous if the mobile sensor is located in the vehicle mobile device and if the spatial area is a local multiple radio cells comprehensive coverage area of a cellular mobile network, and if those signals are detected, the logging in or out of the mobile device or indicate mobile phones in the respective radio cell, which is detected on the basis of the radio cell identifier and the time sequence of the signals in which radio cell, the vehicle at which time.

In this context, preferably based on a geographical assignment of the radio cells to those located in the spatial area Detected traffic routes or checked whether the vehicle moves along one of the traffic routes, in particular with a small, typical of disturbed traffic or large, uniform, typical for free traffic speed moves.

The system according to the invention is advantageously further developed in that the receiving devices are base stations of a cellular mobile radio network, wherein the mobile sensor is a mobile device located in the vehicle or mobile telephone which transmits the signals to the base station, and in that the spatial area comprises a local multiple radio cells Represents supply area of the cellular mobile network, which is supplied by the base stations, and that the center detects and evaluates the received signals from the base stations.

• Die Figur 1 stellt ein Weg-Zeit-Diagramm zur Verkehrssituation dar, welches mit einer herkömmlichen Methode aus vielen Einzelsignalen von stationären oder mobilen Detektoren erstellt wurde;
• Die Figur 2a zeigt einen Kartenausschnitt mit einem geografischen Bereich, der einem Teilersorgungsbereich eines zellulären Mobilfunknetzes entspricht, auf welchen das Verfahren exemplarisch angewendet wird;
• Die Figur 2b zeigt eine Tabelle, die Angaben, insbesondere Funkzellen-Parametern enthält, welche die Zentrale zur Durchführung des Verfahrens verwendet;
• Die Figur 3 gibt ein Weg-Zeit-Diagramm wieder, welches mit dem erfindungsgemäßen Verfahren aus dem Einzelsignal zumindest eines mobilen Detektors erstellt wurde und welches anzeigt, auf welchen Streckenabschnitten und zu welcher Zeit für den Verkehr "Freie Fahrt" herrscht;
• Die Figur 4 zeigt ggf. auftretende Lücken bzw. Schatten in diesem Weg-Zeit-Diagramm, wobei mit dem erfindungsgemäßen Verfahren anhand der Lücken bzw. Schatten erkannt wird, ob und wo eine Verkehrsstörung vorherrscht.

In the following, the invention and the advantages resulting therefrom are described in detail by means of exemplary embodiments, reference being made to the accompanying drawings:

• The FIG. 1 represents a path-time diagram of the traffic situation, which was created by a conventional method from many individual signals from stationary or mobile detectors;
• The FIG. 2a shows a map with a geographical area corresponding to a divisional coverage area of a cellular mobile network to which the method is applied by way of example;
• The FIG. 2b shows a table containing indications, in particular radio cell parameters, which the control center uses to carry out the method;
• The FIG. 3 is a path-time diagram again, which was created with the inventive method from the single signal of at least one mobile detector and which indicates on which sections and at what time for the traffic "free ride"prevails;
• The FIG. 4 shows possibly occurring gaps or shadows in this path-time diagram, it being recognized with the method according to the invention on the basis of the gaps or shadows, whether and where a traffic disturbance prevails.

Based on Fig. 1 Here, first, the conventional method for detecting a traffic jam with conventional sources of traffic data is shown: Here, measurement and reporting data are transmitted as individual signals from stationary and mobile sensors or by human observers (detectors) to a central office. From these many individual signals of the stationary or mobile detectors, the control center then determines the position, length and severity of a traffic jam, and the jam finally presents itself as an extended object in the path-time diagram. As based on the Fig. 2 becomes clear that in the period between 14 h to about 20 h increased traffic on the stretch between 335 km and 350 km has been determined. In the path-time diagram, low speeds are shown as light and dark, resulting in a tongue-shaped object representing the congestion area which changes dynamically over time and extends to the 345 km section of the route. The extension of the congestion or the traffic disruption has in the time of about 19 h, the longest extent, which extends from the kilometer 335-345. In the remaining times the traffic jam was much shorter. Although the conventional methods provide direct information about the presence of traffic congestion, but are relatively expensive to implement because they require in particular a larger number of individual signals and thus sensors or detectors to provide reliable results.

In contrast, it will now be based on the Fig. 2 to 4 the method according to the invention and the system operating thereafter are described in more detail:

In the Fig. 2 a geographical map is shown, which includes the motorway area west of Stuttgart including the motorway junction Leonberg with sections of the A81 and the A8. The last stretch on the A81 in the south to the Leonberger triangle is here to exemplify the section between 335 and 345 km, to which also the path-time diagrams shown here (see Fig.1 such as Fig. 3 and 4 ) and which falls within the spatial range considered here.

This one in the Fig. 2a shown spatial area is covered, inter alia, by a plurality of base stations of the cellular mobile radio network D1, in each case a plurality of mutually adjacent radio cells form a so-called. Radio coverage area LA (Location Area), each associated with a specific identifier, the so-called. LAC (Location Area Code). The radio cells themselves each have their own radio cell identifier Cl (Cell ID). The radio cells are usually supplied by sector antennas, which have a certain opening angle AZM (Azimuth Angle) depending on the requirement. These data or signals and further parameters are managed centrally in the mobile radio network, in particular also to detect the so-called. Logging in and out of mobile stations when they move through the radio coverage areas LA or individual radio cells and thus each from one radio cell to the next must be passed on. This handover (so-called handover or handoff or radio cell change) can only take place without interruption if the mobile radio system constantly updates this data and initiates the necessary handover processes at an early stage.

The invention makes use of these already existing system functionalities in order to realize a very effective evaluation for detecting traffic disruptions. Because it is a special finding of the invention that to solve the problems formulated at the outset, mobile sensors, in particular mobile phones or terminals, are outstandingly suitable for detecting free-flowing traffic and then gaining knowledge about traffic jams and even making predictive statements can.

For mobile sensors, e.g. As "floating cars" are on the way here, the following relationship was recognized, from which to derive assumptions about the traffic flow, which are very likely to apply:

From the fact that a floating car has covered a greater distance at high speed, it immediately follows that there must have been free traffic on the route between sections on high functional road classes (eg motorways or highways), since both detours (the also easily detectable with Floating Cars) as well as traffic disturbances would have led to delays.

But also for mobile phones this statement is correct. Even if the assignment of measurement data to the digital road map does not succeed unambiguously, and even if there are doubts about the assignment of measurement data to cars (as opposed to trains, buses or similar), then: A large amount of traffic (which, incidentally, in the Data of the mobile network according to the market share of the operator) can be detected at a high speed (which is easily recognizable in the data of the mobile network, see table in Fig. 2b ) are only transported on high functional class roads that are relatively easily identifiable.

The mobile network thus generates in clear sequence clear signals for free traffic, which are easily attributable to a digital road map.

As exemplified by in the Fig. 2b 1, a mobile telephone is detected at the time 06:28:32 in the radio coverage area with the identifier LAC = 28939, when entering the radio cell with the identifier Cl = 1222, ie at the northern edge of the radio coverage area LA. Based on the further log-on and log-off events in the following radio cells, which have the identifiers Cl = 1798, 1800, 1563, 3209 and 5417, it is clear that the mobile phone moves to the southern end of the area LA. Due to the short time intervals or the short period of time after which the coverage area LA is already left at 06:32:24, it can be concluded that the mobile phone has moved at a high speed through the area LA, which is finally compensated With the geographical data (digital road map) concludes that the mobile phone must be in a fast-moving vehicle, which must have moved on the highway, namely on the A81 and the A8 to the south.

Based on the evaluation in the table (see Fig. 2b ) data or parameters is clear: The mobile phone can only on the in the 2a marked highways have moved from north to south at a relatively high and steady speed.

Thus, already by tracking and evaluating the signal of a single mobile sensor and by comparison with the cartographic data created the possibility to give a very high probability of a reliable statement about free travel. The signal detected by the mobile sensors thus corresponds to a reliable "free ride" signal.

If one now captures one or preferably several such signals in a path-time diagram, one obtains a representation such as the one shown in FIG Fig. 3 shows. In this, the sections of the route are occupied with a free-drive state for the respective time, provided that the corresponding signals can be detected in the control center. Only if the signals are not present or if there are other signals or parameters which do not indicate "free travel", gaps or shadows form in the path-time diagram.

In the method proposed here, the procedure is virtually the reverse of the conventional method: the traffic jam throws a "shadow" into the free traffic or into the space-time carpet of the free-ride signals (see Fig. 3 ). The absence of signals for free traffic is the reason for the headquarters to report a traffic incident.

Their timing and, finally, their deregistration is also revealed by monitoring the complement of those routes for which clear freedom of movement can be decided. If the decision has been made, the traffic disturbance is indicated by corresponding characters or lines in the path-time diagram (see Fig. 4 ).

Accordingly, a traffic disruption is then registered, if there is a sufficiently large gap in the free-ride signals (s. Fig. 3 ), eg if the average time between individual signals has passed twice on a section of the route, but no signal has arrived. As the gap grows, change messages with corrected length specification are generated. A free-ride signal for the entire affected section leads to the deregistration.

• "10.10.2010 07:11:30, 3 km Stau zwischen Anschlussstelle Mustertal und Kreuz Beispielstadt ab Streckenkilometer 42"

The proposed method can be advantageously designed by adding information about the severity of the disturbance over the course of the traffic message. When registering is usually only the Information on the location and extent of the "shadow" are present, and the message content would be essentially limited to information such as in this example:

• "10.10.2010 07:11:30, 3 km traffic jam between junction Musterertal and cross example city starting from track kilometer 42"

• "10.10.2010 07:19:30, 5 km Stau zwischen Anschlussstelle Mustertal und Kreuz Beispielstadt ab Streckenkilometer 39, Zeitverlust 15 Minuten, Tendenz zunehmend"

Speed measurements in the immediate vicinity, even if they come from the mobile network and can not immediately be clearly assigned to a road, at least make an upper limit to the still possible speed on the recognized as disturbed route. Even a floating car or, in exceptional cases, a stationary sensor detected by a traffic jam, can contribute speed measurements, so that the next message version could be more comprehensive and could be something like:

• "10.10.2010 07:19:30, 5 km traffic jam between junction Musterertal and cross example city from distance kilometer 39, time loss 15 minutes, tendency increasingly"

A method applicable in this context for generating announcements, changes and cancellations is inherent in the DE 199 05 284 A1 and is not the core of this invention.

The inventive method is characterized by determining the typical frequency of the signals for free traffic as a function of distance, time of day and day of the week from historical data. From the frequency response, a minimum time is determined (eg, twice the mean time interval between two signals for free traffic) that must be awaited before a traffic incident can be reported. As a result, the lowest possible logon delay is achieved. You can also set a limit for determine the measured frequency below which the procedure must be deactivated: at low traffic flows, such as at night, a traffic jam will no longer cast shadows and should then, as usual, be detected directly (by means of signals for disturbed traffic).

In summary, therefore, a method is proposed which detects traffic disturbances as space-time complement of spatially complete, localizable, temporally in close succession signals for free traffic, which are preferably generated by mobile sensors, typically mobile phones, but also "floating cars" , and from it traffic information generated for use in information services.

Claims (6)

1. Method for detecting traffic disruptions in a dynamically changing traffic situation in a spatial area (LA), using a mobile telephone network that has radio cells, wherein the mobile telephone network provides signals in a time sequence which indicate a presence of the vehicle in the spatial area (LA) when a mobile telephone device enters and exits the radio cells (LAC), wherein an unimpeded progress of the vehicle along a traffic route (A81, A8) located in the spatial area (LA) is inferred from comparison of the time sequence of the signals with a digital street map, and that, in a manner complementary to the analysis of the time sequence of the signals, an occurrence of traffic disruptions is detected and displayed on the basis of the absence of the signals,
   characterized in that
   typical frequencies of the signals for free traffic flow are ascertained from historical data as a function of route section, time of day, and day of the week, wherein a minimum time that must be waited before reporting a traffic disruption is ascertained from the frequency behavior.
2. Method according to claim 1, characterized in that a repeated absence of the signals to be detected is present during a predefinable time span.
3. Method according to any one of the preceding claims, characterized in that a signaled sign-on and/or sign-off of the vehicle in the spatial region (LAC) takes place.
4. Method according to any one of the preceding claims, characterized in that the spatial region represents a local service area (LA) comprising multiple radio cells of a cellular mobile telephone network, and in that the signals that indicate the sign-on or sign-off of the mobile telephone device in the radio cells are detected, wherein the radio cell identifier (CI) and the time sequence of the signals are used to determine which radio cell the vehicle is located in at any given point in time.
5. Method according to claim 4, characterized in that a geographic association of the radio cells with the traffic routes (A81, A8) located in the spatial area (LA) is used to detect whether the vehicle is moving along one of the traffic routes (A81), in particular at a low speed typical of a traffic jam or at a high, uniform speed.
6. System for detecting traffic disruptions in a dynamically changing traffic situation in a spatial area (LA), wherein the system comprises receiving devices in the form of base stations of a cellular mobile telephone network and a central office connected thereto, wherein the mobile telephone network provides signals in a time sequence which indicate a presence of the vehicle in the spatial area (LA) when a mobile telephone device enters and exits the radio cells (LAC), wherein a continuous progress of the vehicle along a traffic route (A81, A8) located in the spatial area (LA) is inferred from a comparison of the time sequence of the signals with a digital street map, and that the central office, in a manner complementary to the analysis of the time sequence of the signals, detects and prepares for display, or for forwarding to users or systems for further processing, an occurrence of traffic disruptions on the basis of the absence of the signals, characterized in that
   typical frequencies of the signals for free traffic flow are ascertained from historical data as a function of route section, time of day, and day of the week, wherein a minimum time that must be waited before reporting a traffic disruption is ascertained from the frequency behavior.

Images