Method for generating a lane change signal for an ego vehicle and system

Abstract

The invention relates to a method for generating a lane change signal for an ego vehicle. The method according to the invention is characterized by the following process steps: A: Collecting lane change information, describing a time interval (t1, t2) between a lane-changing vehicle (2), performing a lane change on the multi-lane road section (1), and a vehicle (4) ahead of (4) and / or following (5) the lane-changing vehicle (2) on the destination lane (3) of the lane change at the completion of the lane change; B: Detecting the location position of the lane-changing vehicle (2) when collecting lane-changing information; C: Aggregating the lane change information on a processing unit; D: Classification of the road segments (1) into hazard classes (HC1, HC2, HC3, HC4) by the computing unit depending on the respective time intervals (t1, t2) recorded on the respective road segments (1); and E1: Issuing a lane change signal containing a warning to a driver of a manually controlled ego vehicle when the ego vehicle is on a road section (1) of a first hazard class (GC1); and / or E2: Issuing a lane change signal to prevent an automatic lane change for an at least partially automated ego vehicle when the ego vehicle is on a road section (1) of the first hazard class (GK1).

Description

[0001] The invention relates to a method for generating a lane change signal for an ego vehicle of the type defined in more detail in the preamble of claim 1 and to a system for carrying out the method.

[0002] Changing lanes, for example on a motorway, always carries risks. In particular, other road users can be overlooked, meaning that when merging into the adjacent lane, the minimum safe distance to a vehicle ahead or behind is not maintained. An insufficient safety distance increases the risk of an accident if the vehicle ahead brakes sharply after merging. The driver of the following vehicle may also try to close the previous gap by slowing down. This can cause vehicles behind the following vehicle to brake as well, creating a chain reaction of braking maneuvers that propagates against the direction of travel along the motorway. This can contribute to traffic congestion.

[0003] It is therefore desirable to specify means that simplify the execution of lane changes on a multi-lane road section for both manually controlled and at least partially automated vehicles while maintaining road safety.

[0004] The lane change assistance system for autonomous vehicles is known from German patent DE 10 2014 000 843 A1. This document describes the autonomous execution of a lane change as soon as a sufficiently large gap exists in the target lane. The vehicle performing the lane change and the vehicles already in the target lane can communicate with each other. The vehicles can coordinate their operating behavior to specifically create space for the vehicle changing lanes. As an incentive to create a sufficiently large gap, vehicles in the target lane can be credited with a toll bonus.

[0005] German patent DE 10 2020 002 993 A1 discloses a method for generating a lane change signal for a vehicle, whereby data relating to lane changes on a multi-lane road section are collected.

[0006] If more than two automated vehicles are driving in the lane and / or the overtaking lane, the next vehicle is classified as suitable for creating a merging gap for the vehicle by means of a central computer unit, based on which vehicle is closest to the first vehicle.

[0007] The present invention is based on the objective of providing an improved method for generating a lane change signal for an ego vehicle, the implementation of which can increase road safety.

[0008] According to the invention, this problem is solved by a method for generating a lane change signal for an ego vehicle with the features of claim 1. Advantageous embodiments and further developments as well as a system for carrying out the method are described in the dependent claims.

[0009] A generic method for generating a lane change signal for an ego vehicle, wherein data relating to the lane changes of vehicles on a multi-lane road section are collected, is further developed according to the invention by the following method steps: A: Collecting lane change information, describing a time interval between a lane-changing vehicle performing a lane change on the multi-lane road section, and a vehicle preceding and / or following the lane-changing vehicle on the target lane of the lane change at the completion of the lane change; B: Detecting the position of the lane-changing vehicle when collecting lane-changing information and enriching the lane-changing information with an assignment between time interval and road segment of the underlying road network; C: Aggregating the lane change information on a processing unit; D: Classification of road segments into hazard classes by the processing unit depending on the respective time intervals recorded on the respective road segments, whereby a road segment is assigned to a first hazard class if the frequency of falling below the time interval of a specified minimum distance is greater than a specified frequency threshold and the road segment is assigned to a second hazard class if the frequency is less than the frequency threshold; and E1: Issuing a lane change signal containing a warning to a driver of a manually controlled ego vehicle when the ego vehicle is on a road section of the first hazard class; and / or E2: Issuing a lane change signal to prevent an automatic lane change for an at least partially automated ego vehicle when the ego vehicle is on a road section of the first hazard class.

[0010] The method according to the invention thus provides for the classification of individual road sections of the underlying road network into hazard classes depending on the distance between vehicles after a lane change. This makes it possible to identify road sections of multi-lane roads where lane changes frequently occur with a critical safety distance being breached. A time interval is used as a measure for evaluating the distance between vehicles. The time interval describes the duration that the vehicle changing lanes requires, after completing the lane change, to reach the location of the vehicle ahead in the destination lane at the time the lane change is completed.The time interval describes, for the following vehicle, the duration it takes for the following vehicle to reach the position the lane-changing vehicle occupies at the end of the lane change. The time interval is thus determined by the spatial distance between the respective vehicles, taking into account their respective speeds. "Completing" the lane change means that the lane-changing vehicle has fully reached the destination lane. This occurs when all parts of the lane-changing vehicle have crossed the lane markings separating the exit and destination lanes.

[0011] The method according to the invention therefore assesses the risk of lane changes only at their end and not at their beginning or during the execution of the respective lane change.

[0012] Various methods can be used to collect lane change information, which will be discussed in more detail later.

[0013] In step C, the lane change information is aggregated on a processing unit. This can be an in-vehicle processing unit. Processing the lane change information, the unit then classifies the road segments into the aforementioned hazard classes. A distinction is made between at least two hazard classes. To classify road segments into hazard classes, the system checks how frequently the time intervals described by the lane change information fall below the defined minimum distance on the respective road segment. Generally, it is also conceivable to define different minimum distances and frequencies, thus enabling a division into more than two hazard classes.

[0014] Generally, an absolute or relative frequency is suitable as a measure of frequency, such as at least 1000 violations of the minimum distance in total, at least 1000 violations of the minimum distance within the last 3 weeks, or a ratio such as 75% of the observed lane changes are characterized by a violation of the minimum distance in at least one direction or in both directions, while only 25% of the observed lane changes forward or backward or forward and backward maintained the minimum distance.

[0015] The ego-vehicle can receive the classification of road segments into hazard classes from the processing unit and, taking its own position into account, determine whether it is on a road segment of the first or second hazard class. Alternatively, a processing unit implemented within the ego-vehicle can be used. If the ego-vehicle detects that it is currently on a road segment of the first hazard class, a warning message can be issued to the driver if it is being driven manually. However, if the ego-vehicle is at least partially automated or even autonomously controlled, automated or autonomous lane changes can be omitted even when it is on a road segment of the first hazard class.By issuing appropriate warnings to the driver, they can be made aware that lane changes are particularly dangerous on the current section of road. This prompts the driver to more carefully check the distances between vehicles in the target lane before manually changing lanes, or to refrain from manual lane changes altogether. A vehicle that is at least partially automated or autonomously controlled, on the other hand, will not automatically perform lane changes, thus preventing the risk of falling below the required minimum distances between vehicles in the target lane. Road safety can therefore be improved.

[0016] More modern vehicles typically have more sensitive sensors, enabling even more precise environmental perception than older vehicles. They also usually incorporate modern processing units, which, due to their increased performance, allow for faster processing of sensor data. Because more modern vehicles are characterized by improved environmental perception, higher hazard classes for road sections can be tolerated. Thus, while warnings are issued or semi-automated lane changes are prevented for a manually controlled ego vehicle or a comparatively old, at least partially automated, ego vehicle, semi-automated lane changes can be permitted for a modern vehicle at a given hazard class.

[0017] An advantageous further development of the method according to the invention provides that automatic lane changes for the at least partially automated Ego vehicle are only permitted on road sections of the second hazard class. This allows for particularly safe operation of the Ego vehicle, since automatic lane changes are thus prevented even if no information regarding the classification of the road section in question into hazard classes is available. Automatic lane changes are therefore only permitted for those road sections for which a corresponding "safe" classification exists.

[0018] Preferably, the lane change signal is issued in such a way that the time interval between the lane-changing vehicle and the vehicles ahead and behind it is equalized. The ego vehicle itself acts as a lane-changing vehicle when it performs a lane change. The ego vehicle preferably changes lanes in such a way that it positions itself relatively centrally between the vehicles ahead and behind it. This represents a particularly safe way to perform lane changes, as it increases the minimum existing time interval.

[0019] In a manually controlled ego-vehicle, a corresponding warning is issued to the driver. This can be achieved either by timing the warning signal according to the respective time intervals and / or by providing relevant additional information. For example, the indicator light of a blind spot monitoring system integrated into a side mirror can illuminate in a first color, such as red or orange, when the ego-vehicle is on a road section classified as high-risk, and then change to a second color, such as green or blue, when there is a sufficient time gap to the vehicles ahead and behind in the target lane.The driver can thus be shown a suitable time window for making lane changes, so that the ego vehicle can be operated safely despite making a lane change on a comparatively dangerous section of road.

[0020] According to a further advantageous embodiment of the method according to the invention, it is further provided that in step E1 the warning message is issued upon detection of an intention to change lanes, in particular indicated by the detection of a shoulder check by the driver of the ego-vehicle and / or the activation of a turn signal. Thus, warning messages are preferably issued to the driver of a manually controlled ego-vehicle only as needed. This reduces the risk of the driver being distracted by such warning messages. The driver can be detected using interior sensors to identify the intention to change lanes. Visual systems such as cameras or other sensors that enable the generation of depth information, such as radar sensors or other motion sensors, can be used as sensors.Such sensors can be configured to recognize gestures. In particular, they can locate the individual limbs or body parts of the driver inside the vehicle and, by tracking their geometric orientation, infer that a shoulder check is being performed. Additionally or alternatively, the intention to change lanes can also be detected by the activation of the turn signal. A distinction is made between a left and right shoulder check, or between activating the left or right turn signal. The intention to change lanes is therefore not only an indication that the driver intends to manually change lanes, but also which adjacent lane they plan to change to. For example, if the driver is traveling in the ego-vehicle in the middle lane of a three-lane highway, a lane change to the left or right lane is generally possible.

[0021] A further advantageous embodiment of the method according to the invention provides that the computing unit classifies road segments into hazard classes depending on the current traffic flow on the road segment, using traffic-flow-dependent values ​​for the minimum distance and / or the frequency threshold. The accident risk can increase for roads with higher traffic flow. To adequately account for this, road segments can be classified into hazard classes taking the traffic flow into account. For this purpose, different values ​​for the minimum distance and / or the relevant frequency threshold are selected depending on the observed traffic flow. The ego-vehicle can be capable of determining the traffic flow on the road segment it is traveling on itself, particularly in real time.For this purpose, the relative distance to surrounding vehicles can be determined using environmental sensors such as cameras, radar sensors, ultrasonic sensors, laser scanners like LiDAR, and similar devices. A comparatively small relative distance or a comparatively high vehicle density indicates a relatively high traffic flow. Corresponding traffic flow information can also be obtained in the usual way, for example, from a traffic service. Such a service typically uses stationary measurement technology, such as traffic monitoring cameras, to collect data.

[0022] Specific values ​​for the minimum distance or the frequency threshold for at least one of the following types of traffic flow are particularly preferred: free, flowing, stop-and-go, and / or congestion. The terms chosen to characterize traffic flow are particularly common and widely used. Free traffic is characterized by the unimpeded or nearly unimpeded movement of vehicles, usually at or above the applicable speed limit. Traffic density is low.

[0023] In contrast, with flowing traffic, vehicles move continuously, although their speed may be reduced compared to free-flowing traffic. However, the vehicles move in a constant stream without stopping too frequently. Traffic density may be higher.

[0024] Stop-and-go traffic is characterized by vehicles not moving at a constant speed, but rather having to brake and accelerate regularly due to increased traffic density. Frequent changes between brief stops and sudden accelerations occur.

[0025] Congestion, on the other hand, describes a situation in which the flow of traffic is severely restricted. Vehicles move relatively slowly or are stationary for extended periods. There is a high traffic density with a particularly low average speed, which can even temporarily reach zero km / h.

[0026] Free-flowing traffic, smooth-running traffic, stop-and-go traffic, and congestion are therefore the most common types of traffic flow. For each type of traffic flow, specific values ​​can be defined for the minimum distance and / or the frequency threshold.

[0027] A further advantageous embodiment of the method according to the invention provides that the computing unit takes into account infrastructure parameters characterizing the road segment to determine the minimum distance and / or the frequency threshold, in particular in the form of the lane width, the curve radius, the applicable speed limit, and / or the road surface. The hazard class of the respective road segment can thus vary not only with regard to the current traffic flow but also with regard to the aforementioned infrastructure parameters. For example, for road segments with a particularly narrow lane width, a small curve radius, a high speed limit, or a brittle road surface, larger values ​​for the minimum distance or smaller values ​​for the frequency threshold can be considered. The corresponding road segment is therefore more likely to be assigned to the first hazard class.Accordingly, the ego vehicle is controlled appropriately in each dangerous situation.

[0028] According to a further advantageous embodiment of the method according to the invention, the processing unit also takes into account the ratio of the time interval of the lane-changing vehicle to the vehicle in front and the time interval of the lane-changing vehicle to the vehicle behind as an additional parameter for classifying road segments into hazard classes. Road segments are preferably considered safe if the time interval of the lane-changing vehicle to the vehicle in front and the vehicle behind is comparatively equal. If, on the other hand, the two time intervals differ too greatly, this means that the lane-changing vehicle is maintaining too small a time interval to either the vehicle in front or the vehicle behind.Even if the relevant minimum distance for the road section is maintained, this can still increase the risk of an accident, so the road section should and can be classified as the first hazard class.

[0029] A further advantageous embodiment of the method according to the invention provides that a central computing unit acts as a processing unit, which communicates, in particular, lane change information and / or the classification of road segments into hazard classes between the vehicles on the multi-lane road segment. Such a central computing unit can also be referred to as a cloud server or backend. The vehicles in a fleet can communicate wirelessly with the central computing unit. Each vehicle can be equipped with a telecommunications unit that allows for an internet connection, for example, via mobile network. Lane change information is thus aggregated centrally, which enables a reliable classification of the road segments of the road network into their respective hazard classes within a short time.The respective information can then be distributed to the vehicles in the fleet and used effectively there.

[0030] According to a further advantageous embodiment of the method according to the invention, lane change information is also generated by vehicles in a fleet, a drone monitoring a particular road segment, and / or by stationary traffic monitoring cameras installed on the respective road segment. This allows various data sources to be used to generate lane change information. As already mentioned, vehicles can be equipped with suitable sensors that enable the detection of the respective time intervals between vehicles in a target lane when a lane-changing vehicle performs a lane change. A lane-changing vehicle can also be a vehicle from the fleet, a self-driving vehicle, or even a vehicle not belonging to the fleet.Lane change information can therefore be collected by vehicles involved in the lane change or by vehicles merely observing the lane change.

[0031] To determine their position, vehicles can be equipped with common positioning devices such as navigation systems. Such a navigation system is typically able to determine its position on Earth based on a signal transmitted by a global navigation satellite system. Well-known systems include GPS, Galileo, and similar systems. By comparing the vehicle's position with a digital road map, the current road segment of the road network being traveled by the vehicle can be determined. This allows for the correlation of time intervals with road segments, which in turn enables the processing unit to classify the road into hazard categories. The vehicles in the fleet may also have their own in-vehicle processing units.A central computing unit, such as the aforementioned cloud server, can receive lane change information from the vehicles in the fleet and distribute it to them. This allows the assignment of hazard classes to be performed either in the computing units within the vehicles themselves or centrally.

[0032] In addition to vehicles, drones or stationary sensors can also be used to generate lane change information. These drones can be multicopters or fixed-wing aircraft. Such a drone can also be referred to as a UAV (Unmanned Aerial Vehicle). The drone is equipped with appropriate environmental sensors that allow it to detect the time intervals between vehicles changing lanes and the vehicles ahead and / or behind them in the target lane. These sensors include, in particular, camera-based systems, radar, and laser scanners such as LiDAR. Furthermore, the drone can also be equipped with appropriate navigation satellite-based positioning systems.

[0033] When using stationary traffic monitoring cameras, assigning them to specific road segments is particularly easy, as the cameras themselves do not move. Information is available for each camera indicating which road segment it is monitoring.

[0034] In a generic system comprising vehicles and at least one computing unit, the vehicles are configured, according to the invention, to collect and / or receive lane change information enriched with the current position of a respective vehicle on the multi-lane road section, wherein the computing unit is configured to perform steps C and D of a method described above, and the vehicles are configured to perform at least step E1 or E2 of a method described above. When using exclusively in-vehicle computing units, such a computing unit also performs, in particular, steps C, D, and E1 or E2. If the vehicles themselves collect said lane change information, steps A and B are also performed by the respective vehicle or the in-vehicle computing unit.When using drones and / or stationary traffic monitoring cameras, steps A and B can be omitted by the vehicles. Furthermore, if a central computing unit, such as the aforementioned cloud server, is used, steps C and D can be performed externally by the vehicles.

[0035] This can include all types of road vehicles, such as cars, trucks, vans, buses, and the like. The vehicles can be manually and / or at least semi-automatically controlled. Manually controlled vehicles have appropriate output devices for issuing warnings to the driver. Warnings can be issued via various transmission methods, in particular acoustic, visual, and / or haptic. Acoustic warnings can be issued via bells, sirens, loudspeakers, or similar devices; visual warnings via lights, displays, and the like; and haptic warnings via vibration or impacts on a structure touched by the driver.

[0036] Further advantageous embodiments of the inventive method for generating a lane change signal for an ego vehicle and of the inventive system also result from the exemplary embodiments, which are described in more detail below with reference to the figures.

[0037] This shows: Fig. 1 A schematic representation of a vehicle changing lanes, performing a lane change on a multi-lane road section from a bird's-eye view; Fig. 2. A diagram showing the classification of road sections into hazard classes depending on the time interval between the lane-changing vehicle and vehicles ahead and behind; and Fig. 3 a diagram showing the classification according to Fig. 2 with an adapted system behavior of the lane-changing vehicle.

[0038] Fig. Figure 1 shows a top view of a typical traffic situation in which a lane-changing vehicle 2 performs a lane change from an exit lane 6 of a multi-lane road section 1 to a destination lane 3. A vehicle 4 is located ahead of it and a vehicle 5 is following it. Vehicles 2, 4, and 5 are traveling in the direction of travel F. According to the invention, after completion of the lane change, a time interval t1 between the lane-changing vehicle 2 and the preceding vehicle 4, as well as a time interval t2 between the lane-changing vehicle 2 and the following vehicle 5, are determined. The respective time intervals t1 and t2 describe the duration that each vehicle 2 and 5 requires to reach or assume the position of the preceding vehicle 4 and 5, respectively.For this purpose, distances between vehicles 2, 4, and 5 can be measured, and the time intervals t1 and t2 can be calculated based on their respective speeds. Together with location information, these time intervals t1 and t2 are aggregated as lane change information on a processing unit. For example, the vehicles 2, 4, and 5 shown, or even a vehicle not involved in the lane change (not shown), a drone monitoring the respective road segment 1, and / or stationary infrastructure such as traffic monitoring cameras can be used to generate the respective lane change information.

[0039] For aggregation, either an in-vehicle computing unit or a central computing facility such as a cloud server can be used. Respective road segments 1 are divided into the... Fig. The hazards are classified into the 2 or 3 hazard classes GK1 to GK4 shown. The procedure is based on... Fig. 2 explained.

[0040] The x-axis of the diagram shows the time interval t1 to the vehicle 4 ahead, and the y-axis shows the time interval t2 to the vehicle 5 behind. Various measurement points, corresponding to lane changes, are plotted on the diagram. For each lane change, the measured time intervals t1 and t2 are compared to a minimum distance MA. Road segment 1 is assigned to the first hazard class GK1 if the frequency of the time interval t1 and t2 falling below the defined minimum distance MA exceeds a specified frequency threshold. The frequency is indicated by the number of points in the respective diagram segment. Conversely, if the frequency is less than the frequency threshold, road segment 1 is assigned to the second hazard class GK2.

[0041] At the in Fig. 2 and Fig. In the embodiment shown in Figure 3, the diagram is divided into four hazard classes GK1, GK2, GK3, and GK4. The third and fourth hazard classes, GK3 and GK4, can also be considered as the first hazard class, GK1, according to the invention, since at least one of the time intervals t1 and t2 falls below the defined minimum distance MA. For hazard class GK1 shown in the diagram, both time intervals t1 and t2 fall below the minimum distance MA.

[0042] In general, it is conceivable to provide for more than two or four hazard classes GK1 to GK4, which is indicated by further minimum distances MA*.

[0043] According to the invention, a warning message is issued to the driver of a manually controlled ego vehicle when the ego vehicle is on a road section 1 of the first hazard class GK1. This prompts the driver to monitor surrounding traffic more carefully when performing or planning lane changes, or to refrain from performing the lane change altogether. For an ego vehicle that is at least partially automated, the need for at least partially automated lane changes on road sections 1 of the first hazard class GK1 is eliminated.

[0044] Fig. Figure 3 illustrates an advantageous way to operate the Ego vehicle. In the Fig.In the embodiment shown in Figure 3, lane changes are performed in such a way that the time intervals t1 and t2 are aligned as closely as possible. This prevents, where possible, any shortfall below a minimum distance MA, thus increasing road safety. The diagram area representing the second hazard class, GK2, shows a line running diagonally from the bottom left to the top right. Measurement points lying on this line represent lane changes where the two time intervals, t1 and t2, are equal.

Claims

[1] Method for generating a lane change signal for an ego vehicle, wherein data relating to the lane changes of vehicles on a multi-lane road section (1), characterized by the following procedural steps: A: Collecting lane change information, describing a time interval (t1, t2) between a lane-changing vehicle (2), performing a lane change on the multi-lane road section (1), and a vehicle (4) ahead of (4) and / or following (5) the lane-changing vehicle (2) on the destination lane (3) of the lane change at the completion of the lane change; B: Detecting the position of the lane-changing vehicle (2) when acquiring the lane-changing information and enriching the lane-changing information with an assignment between time interval (t1, t2) and road segment (1) of the underlying road network; C: Aggregating the lane change information on a processing unit; D: Classifying the road segments (1) into hazard classes (HC1, HC2, HC3, HC4) by the processing unit depending on the respective time intervals (t1, t2) recorded on the respective road segments (1), wherein a road segment (1) is assigned to a first hazard class (HC1) if the frequency of falling below the time interval (t1, t2) of a specified minimum distance (MV) is greater than a specified frequency threshold and the road segment (1) is assigned to a second hazard class (HC2) if the frequency is less than the frequency threshold; and E1: Issuing a lane change signal containing a warning to a driver of a manually controlled ego vehicle when the ego vehicle is on a road section (1) of the first hazard class (GC1); and / or E2: Issuing a lane change signal to prevent an automatic lane change for an at least partially automated ego vehicle when the ego vehicle is on a road section (1) of the first hazard class (GK1). [2] Method according to claim 1, characterized by , that the automatic lane change for the at least partially automated Ego vehicle is permitted only on road sections (1) of the second hazard class (GK2). [3] Method according to claim 1 or 2, characterized by , that in step E1 the warning message is issued when an intention to change lanes is detected, in particular indicated by the detection of a shoulder check by the driver of the Ego vehicle and / or the activation of a turn signal. [4] Method according to any one of claims 1 to 3, characterized by, that the computing unit classifies road segments (1) into hazard classes (GK1, GK2, GK3, GK4) depending on the current traffic flow on the road segment (1), using traffic flow-dependent values ​​for the minimum distance (MA) and / or the frequency threshold. [5] Method according to claim 4, characterized by , that specific values ​​for the minimum distance (MA) and / or the frequency threshold are used for at least one of the following types of traffic flow: free, flowing, stop-and-go, and / or congestion. [6] Method according to any one of claims 1 to 5, characterized by , that the computing unit takes into account infrastructure parameters characterizing the road segment (1) for determining the level of the minimum distance (MA) and / or the frequency threshold, in particular in the form of the lane width, the curve radius, the applicable speed limit and / or the road surface. [7] Method according to any one of claims 1 to 6, characterized by , that the computing unit takes into account the ratio of the time interval (t1) of the lane-changing vehicle (2) to the vehicle ahead (4) and the time interval (t2) of the lane-changing vehicle (2) to the following vehicle (5) as an additional parameter for classifying the road sections (1) into hazard classes (GK1, GK2, GK3, GK4). [8] Method according to any one of claims 1 to 7, characterized by , that a central computing facility acts as a computing unit which communicates in particular lane change information and / or the classification of road sections (1) into hazard classes (GK1, GK2, GK3, GK4) between the vehicles on the multi-lane road section (1). [9] Method according to any one of claims 1 to 8, characterized by, that lane change information is generated by the vehicles of a vehicle fleet, a drone monitoring a respective road section (1) and / or by stationary traffic observation cameras installed on the respective road section (1). [10] System comprising vehicles and at least one computing unit, characterized by , that the vehicles are equipped to collect and / or receive lane change information enriched with a location position of a respective vehicle on the multi-lane road section (1), wherein the computing unit is equipped to perform steps C and D of a method according to any one of claims 1 to 9 and the vehicles are equipped to perform at least step E1 or E2 of a method according to any one of claims 1 to 9.

Images