ASSIGNMENT OF TRAFFIC LIGHT TO CORRESPONDING LANES

Description

[0001] The invention relates to a method for assigning traffic lights to the corresponding lanes of a roadway and to the use of the method in a vehicle.

[0002] When a motor vehicle is driven autonomously or automatically, it must cope with complex traffic light situations at intersections. This requires automatic detection of the traffic lights present at an intersection. This detection must also include the assignment of the traffic lights to the respective lanes for which the traffic light signals are responsible. Such automatic detection of traffic lights at intersections is not only applicable to autonomous driving, but can also be used in assisted or semi-autonomous driving of a motor vehicle to assist the driver in operating the vehicle.

[0003] Automatic detection of traffic lights at an intersection during automated driving pursues the following goals: to recommend to the driver a possible speed to reach the traffic light, to stop automatically before the traffic light, to warn the driver if the traffic light is accidentally exceeded, and to give the driver a warning to continue driving when the traffic light turns green again.

[0004] To implement these examples, we attempt to observe the traffic light using the vehicle's front camera. The following measurement task must be accomplished: The camera detects the traffic light in advance and measures its position and orientation. The camera detects the lanes and markings on the ground, thus determining the assignment of the traffic light to the lanes. The camera continuously monitors the status of the relevant traffic light and forwards this information to the appropriate function.

[0005] In reality, however, the assignment of traffic lights to lanes is very difficult to determine due to the diverse and complex intersection situations, which leads to high error rates in the assignment of traffic lights to lanes.

[0006] The document DE 10 2014 205 953 A1 relates to a method for analyzing a traffic environment situation of a vehicle with a navigation system and a sensor system, wherein the navigation system determines a vehicle position and the sensor system records movement data of the vehicle and determines a movement behavior of the vehicle in the form of movement trajectories of the vehicle, wherein the vehicle position and the movement trajectories of the vehicle enable a conclusion to be drawn about the presence and / or the type of at least one feature of the environment, wherein a sensor of the sensor system detects the feature and stores at least one item of additional information about the feature.

[0007] The document DE 10 2013 220 662 A1 describes a method for detecting traffic situations when operating a vehicle on a road with several lanes, comprising the following features: With the help of a vehicle-mounted sensor system, various environmental data relating to a status of a traffic signal, a road boundary, a lane boundary, and at least one other marking on the road are optically recorded, and using the environmental data recorded by the sensor system, a relationship is established between the status of the vehicle, which includes at least the information on the lane used by the vehicle, and the status of the traffic signal valid for this vehicle status.

[0008] The subsequently published document DE 10 2018 004 667 A1 relates to a method for determining the direction of travel of a traffic light system. The traffic light system is detected by a camera system mounted on a vehicle, and the detected information is sent to an external computer unit and evaluated by this unit. The information from the detected traffic light system captured by the camera system is correlated with information stored in the external computer unit, and the direction of travel is determined based on the correlation.

[0009] The publication DE 10 2016 217 558 A1 relates to a method for assigning a traffic light to a lane, comparing detected phase sequences with expected phase sequences. The publication also relates to an associated vehicle and an associated storage medium.

[0010] The document US 20140185880 A1 describes a method for determining the three-dimensional position of a traffic light signal. The method receives a plurality of images, each of which is provided with a geographical position and directional information. From the images, those belonging to the geographical surroundings of an intersection are selected, with red, yellow, and green objects being identified in each selected image. Subsequently, associated red, yellow, and green objects are identified in successive images as a function of the geographical position and directional information. Based on the identification of the associated objects in the successive images, the three-dimensional arrangement of the traffic light signals is determined, which is then subsequently stored for further use.Furthermore, a lane assigned to a traffic light signal can be determined by comparing the three-dimensional arrangement of the traffic lights with a map of the lanes. In this way, a three-dimensional map including the three-dimensional arrangement of the traffic lights can be generated and saved.

[0011] US 20160282879 A1 describes a computer-readable map for use in an autonomous vehicle. The detailed map comprises a plurality of lane segments with a plurality of lane connections. Each of the lane connections can extend between two lane segments across an intersection, with traffic light signals being associated with each lane connection. The map is used in an autonomous vehicle to process sensor information about the autonomous vehicle's surroundings based on the map.

[0012] The invention is therefore based on the object of providing a method for assigning traffic lights to lanes and a use of the method with an improved assignment of the traffic lights to the lanes.

[0013] This object is achieved by a method having the features of claim 1 and by the use of the method having the features of claim 5. Preferred embodiments are the subject of the subclaims.

[0014] The method according to the invention for assigning traffic lights of a traffic light system of a road section of an intersection controlled by this traffic light system to the lanes of the road section, wherein the road section has at least one lane, the lanes of the road section have a common direction of travel, and a plurality of vehicles forming a swarm travel on the road section, each vehicle having a front camera, comprises the steps: Detection of an upcoming traffic light by each vehicle in the swarm that is driving on the road section and moving towards the traffic light using the camera, Continuous recording of the traffic light states of the traffic light system and the lane being traveled on, as well as the vehicle speed and vehicle position by each vehicle using the camera of each vehicle, whereby the recording begins at a predetermined distance from the traffic light and continues until the vehicle passes the traffic light, Transmission of the data recorded by each vehicle (traffic light states, lane being traveled on, position and vehicle speed) to a backend computer, Performing the following evaluations of the transmitted data by the backend computer for each vehicle in the swarm: Determining the detected traffic light on a map, Detecting the movement of the vehicle through the road section,Determining the traffic light states of the traffic lights as the vehicle passes through the traffic light system, summarizing the evaluations of the data from each vehicle in the swarm into a respective data set relating to the road section, wherein the data set includes the traffic light states of the traffic lights in the road section limited to red and green, the lane of the road section traveled by the vehicle, and the driving behavior of the vehicle at the traffic lights, namely whether the vehicle has passed through or stopped, and creating an assignment table in the backend computer, accessible by vehicles, in the form of a matrix for the detected traffic light system from the data sets of the vehicle swarm, wherein the assignment table assigns the traffic lights of the traffic light system to the lanes of the road section, the assignment table listing the traffic lights of the traffic light system with their traffic light states, wherein the traffic light states are limited to red and green,and the allocation table lists the lanes of the road section, with the lanes having the binary values ​​for passing through and stopping as a function of the traffic light states.

[0015] The method according to the invention therefore determines the lane assignment of traffic lights using swarm data from a large number of test vehicles. Driver behavior and observations of traffic light conditions are compared in order to determine the correct assignment of traffic lights to lanes for each road section, especially an intersection.

[0016] Preferably, the predetermined distance is 500m, from which the detection of the traffic light states of the traffic light system begins, whereby the predetermined distance may be less depending on the application, for example if several intersections are arranged in quick succession.

[0017] Preferably, the allocation table is stored in a database for each traffic light system, so that the allocation table can be accessed by other vehicles.

[0018] The assignment table lists the traffic lights of the traffic light system along with their states, with the states limited to red and green. By limiting the states to the two relevant states, which indicate whether the light can be passed or whether a stop is required, the states can be expressed in the assignment table using a single bit.

[0019] Furthermore, the lanes are listed in the allocation table, with the lanes having the binary values ​​for passing through and for stopping as a function of the traffic light states.

[0020] Furthermore, the vehicles driving on the roadway section are preferably manually controlled to create the data sets. The "swarm" of test vehicles used to determine the assignment tables is preferably manually controlled so that the assignment of traffic lights to lanes can be retrieved for later autonomous driving at key intersections, ensuring safe autonomous driving.

[0021] When using the method according to the invention for assigning traffic lights of a traffic light system of a road section controlled by this traffic light system to the lanes of the road section according to the method explained above in a vehicle traveling on the road section, the vehicle determines the position of the traffic light system controlling the road section and, based on this position, requests the assignment table of the traffic light system in question from a database and uses this table to support the driver in an assistance system.

[0022] Preferably, the requested assignment table is evaluated in the vehicle together with current vehicle data, wherein the current vehicle data particularly relates to the current lane departure, turn signal settings and / or navigation routes.

[0023] A preferred embodiment of the invention is explained below with reference to the drawings. Fig. 1 an intersection situation with appropriate traffic lights, and Fig. 2 a detailed representation of a traffic light situation for three lanes at an intersection.

[0024] In the Fig. 1 A typical intersection situation is described, which is used to explain the procedure. An intersection K is formed by the meeting of the two carriageways Fb1 and Fb2. The first carriageway Fb1 comprises a total of six lanes F11 to F16, with three lanes F11 to F13 in the Fig. 1run upwards and the other three lanes F14 to F16 run in the opposite direction. The second lane Fb2, which runs perpendicular to the first lane Fb1, comprises two lanes F21 and F22, which run in the opposite direction. The traffic situation at intersection K with the two lanes Fb1 and Fb2 is regulated by four interconnected traffic lights A11, A12, A21 and A22, with traffic lights A11 being responsible for lanes F11 to F13, A12 for lanes F14 to F16, A21 for lane F21 and A22 for lane F22. Furthermore, Fig. 1 For the sake of completeness, pedestrian crossings are drawn in the form of zebra crossings Z.

[0025] The Fig. 1The traffic light systems A11, A12, A21, and A22, which are only symbolically shown, can comprise multiple traffic lights with specific meanings depending on the intersection type and can, for example, be a function of the number of lanes. For example, the traffic light systems A11 and A12 on carriageway Fb1, with its three lanes F11 to F16, regulate driving behavior in each direction: driving straight ahead, turning left, and turning right. These traffic light systems A11 and A12 typically consist of multiple lights, each responsible for the respective driving behavior. In other words, the traffic light systems A11 and A12 comprise separate traffic lights for driving straight ahead, turning left, and turning right.

[0026] This is in the Fig. 2which shows in detail the traffic situation at intersection K with regard to traffic light system A11 with zebra crossing Z and the lanes F11, F12 and F13 controlled by it. The right partial traffic light system AR and the left partial traffic light system AL of traffic light system A11 are used to control traffic on the aforementioned lanes F11, F12 and F13. The right partial traffic light system AR regulates both straight-ahead traffic on the three lanes F11, F12 and F13 with traffic light AR1 and right-turn traffic on the right lane F11 with traffic light AR2. Furthermore, the left partial traffic light system AL also regulates straight-ahead traffic on the three lanes F11, F12 and F13 with traffic light AL1 and left-turn traffic on the left lane F13 with traffic light AL2.

[0027] The procedure includes the following steps: a) The vehicle detects a traffic light system in advance and the status of the individual lights, i.e. green, yellow, red, red / yellow, and detects which lane it is in. This data is continuously recorded until the vehicle has passed through the traffic light. The speed is also recorded to determine whether the vehicle stopped at the traffic light or was able to pass through without stopping. In real environments, there is also the possibility that the traffic light status is not detected correctly. This can occur due to poor visibility conditions (glare, dirt, etc.), obscuring the traffic light (e.g. trucks in the line of sight) or traffic lights outside the field of view or aperture angle of the camera. In these cases, only the detected traffic light states are reported - missing states are not known to the vehicle and therefore cannot be communicated. b) The recorded data is sent to a backend system.c) In the backend the following is done: i) Referencing the detected traffic light system to a map ii) Detecting the movement of the vehicle through the intersection iii) Determining the traffic light status when crossing the intersection .

[0028] For those in Fig. 2 For the traffic light system A11 shown, the following table shows exemplary data records as they are transmitted to the backend computer or are available in the backend computer for processing: Table 1: Traffic light right turn AR2 Traffic lights without arrow AR1 and AL1 Traffic light left turn AL2 Track taken Passed through (1) / Stopped (0) red red green F13 1 red 0 green F12 0 red green red F11 1 red green red F12 1 red green red F13 1

[0029] The following applies to driving behavior at traffic lights: Passing = "1", Stopping = "0".

[0030] The above data sets transmitted to the backend computer contain information about the traffic light states, the lane occupied by the vehicle, and the behavior at the intersection, i.e., whether the vehicle is stationary or passing through. For example, the first line indicates that a vehicle passed through in the left lane F13, with the traffic light AL2 indicating a left turn being green, while all other traffic lights were red. This indicates that the traffic light AL2 is associated with the left lane F13. e) Assuming that the intersection will only be crossed when the light is green, an assignment to a possible lane can now be made. From the example above, I can see that driving is permitted in lane F13 if the traffic light on the left AL2 is green. However, not in lane F12. Information on lane F11 is not yet available. Furthermore, driving is permitted in all lanes if the traffic light without the arrow AR1, AL1 is green. f) Depending on the traffic light system, the traffic light display is ambiguous. For example, both the straight-ahead and the left exit are green in the traffic light system. However, the vehicle only chooses the left exit. This does not lead to any functional impairments, as this is resolved using an assignment table, which must be recorded for each traffic light system. The recording takes place over several passes, which are recorded by the "vehicle swarm."Each vehicle determines the status in the form of a Boolean variable with the values ​​"0" or "1" for the lane at the corresponding traffic light status.

[0031] Possible traffic lights that are not recognized by the vehicles and therefore not reported are not taken into account for the expansion of the table and do not contribute to the information structure - but also do not contribute to fault conditions and can therefore be tolerated.

[0032] From the data of the "swarm fleet" transmitted to the backend, which are shown in Table 1, an allocation table in the form of a matrix is ​​created for the intersection section concerned by the traffic light system A11 and is shown as an example in Table 2. The allocation tables for the Fig. 1 The other intersection sections shown concerning the traffic lights A12, A21 and A22 look accordingly. Table 2: Traffic light right AR2 Traffic lights without arrow AR1, AL1 Traffic light left AL2 Lane F11 Lane F12 Lane F13 (1=green, 0=red) (1=green, 0=red) (1=green, 0=red) (1 = Drive, 0 = Stop) (1 = Drive, 0 = Stop) (1 = Drive, 0 = Stop) 0 0 1 0 0 1 0 1 0 1 1 1 0 1 1 1 1 1 1 0 0 1 0 0 1 0 1 1 0 1 1 1 0 1 1 1 1 1 1 1 1 1 g) Once this assignment has been determined, it is stored in a database in the form of an assignment table for each traffic light and transmitted to the vehicles, e.g., as a map layer. h) If the car now approaches the traffic light again, the possible free lanes can be determined using the assignment table. i) This situation is then evaluated by the functions along with the current lane offset, turn signal settings, navigation routes, etc.

[0033] A further improvement could be achieved if the entry in the table could be determined directly in the vehicle. While this would require extended logic in the vehicle, the data volume could be significantly reduced, as only the path status in relation to the traffic light status needs to be transmitted. List of reference symbols

[0034] KIntersection Fb1Lane 1 of carriageway 1 Fb2Lane 2 of carriageway 1 F11Lane 1 of carriageway 1 F12Lane 2 of carriageway 1 F13Lane 3 of carriageway 1 F14Lane 4 of carriageway 1 F15Lane 5 of carriageway 1 F16Lane 6 of carriageway 1 F21Lane 1 of carriageway 2 F22Lane 2 of carriageway 2 A11Traffic lights for lanes F11 to F13 A12Traffic lights for lanes F14 to F16 A21Traffic lights for lane F21 A22Traffic lights for lane F22 ZZebra crossing ARPartial traffic lights on the right AR1Traffic light for straight-ahead travel AR2Traffic light for right-turning ALPartial traffic lights on the left AL1Traffic light for straight-ahead travel AL2Traffic light for left-turning

Claims

1. Method for allocating traffic lights (AR1, AR2, AL1, AL2) of a traffic light system (A11) of a roadway portion (Fb1) of an intersection (K) that is controlled by this traffic light system (A11) to the lanes (F11, F12, F13) of the roadway portion (Fb1), wherein the roadway portion (Fb1) has at least one lane (F11, F12, F13), the lanes (F11, F12, F13) of the roadway portion (Fb1) have a common direction of travel, and a plurality of vehicles forming a swarm drive on the roadway portion (Fb1), wherein each vehicle has a front camera, the method comprising the following steps: - each vehicle of the swarm that drives on the roadway portion (Fb1) and moves toward the traffic light system (A11) detecting a traffic light system (A11) ahead by means of the camera, - the camera of each vehicle continuously recording the traffic light states of the traffic lights (AR1, AR2, AL1, AL2) of the traffic light system (A11) and the lane (F11, F12, F13) being driven on, and each vehicle recording the vehicle speed and the vehicle position, wherein the recording begins at a predetermined distance from the traffic light system (A11) and continues until the traffic light system (A11) is passed, - transmitting the data recorded by each vehicle, traffic light states, driven lane, position and vehicle speed, to a backend computer, - carrying out the following evaluations of the transmitted data for the vehicles of the swarm by means of the backend computer: - identifying the detected traffic light system (A11) on a map, - detecting the movement of the vehicle through the roadway portion (Fb1), - determining the traffic light states of the traffic lights (AR1, AR2, AL1, AL2) when the vehicle passes the traffic light system (A11), summarizing the evaluations of the data of each vehicle in the swarm into a relevant data set relating to the roadway portion, wherein the data set comprises the traffic light state, limited to red and green, of the traffic lights (AR1, AR2, AL1, AL2) of the roadway portion, the lane on the roadway portion that was driven on by the relevant vehicle, and the driving behavior of the vehicle at the traffic lights (AR1, AR2, AL1, AL2), namely whether the vehicle drove through or stopped, and - creating an allocation table in the backend computer, in the form of a matrix for the detected traffic light system (A11), from the data sets of the vehicle swarm, which allocation table can be accessed by vehicles, wherein the allocation table allocates the traffic lights (AR1, AR2, AL1, AL2) of the traffic light system (A11) to the lanes (F11, F12, F13) of the roadway portion (Fb1), - in the allocation table, the traffic lights (AR1, AR2, AL1, AL2) of the traffic light system (A11) are listed together with the traffic light states, wherein the traffic light states are limited to red and green, and - in the allocation table, the lanes (F11, F12, F13) of the roadway portion (Fb1) are listed, wherein the lanes (F11, F12, F13) have the binary values for driving through and for stopping as a function of the traffic light states.

2. Method according to claim 1, wherein the predetermined distance is 500 m or less.

3. Method according to claim 1 or claim 2, wherein the allocation table for each traffic light system (A11) is stored in a database.

4. Method according to any of the preceding claims, wherein the vehicles driving on the roadway portion (Fb1) are manually controlled to create the data sets.

5. Use of the method for allocating traffic lights of a traffic light system (A11) of a roadway portion (Fb1) controlled by this traffic light system (A11) to the lanes (F11, F12, F13) of the roadway portion (Fb1) according to any of the preceding claims in a vehicle driving on the roadway portion (Fb1), wherein the vehicle determines the position of the traffic light system (A11) controlling the roadway portion (Fb1) and, based on this position, requests the allocation table of the relevant traffic light system (A11) from a database and uses the allocation table to support the driver in an assistance system.

6. Use of the method according to claim 5, wherein the requested allocation table is evaluated in the vehicle together with current vehicle data.

7. Use of the method according to claim 5 or claim 6, wherein the current vehicle data relate to the current lane deviation, turn signal settings and / or navigation routes.