The association between traffic lights and their respective lanes.

Abstract

Association of traffic lights with their respective lanes. A method for associating traffic lights of a traffic signal facility with lanes of a road segment controlled by the traffic signal facility, wherein the road segment has at least one lane, the lanes of the road segment share a common direction of travel, and multiple vehicles travel on the road segment, the method comprising the following steps: - identifying an upcoming traffic signal facility; - detecting the traffic signal status of the traffic lights of the traffic signal facility, the lane on which the vehicles are traveling, the vehicle speed, and the vehicle position; - transmitting data recorded by each vehicle in the form of data sets to a background computer; - performing the following analyses on each vehicle-specific data set: - determining the identified traffic signal facility on a map; - identifying the movement of vehicles through the road segment; - determining the traffic signal status when the vehicles pass the traffic signal facility; and - creating an association table.

Description

TECHNICAL FIELD

[0001] The invention relates to a method for associating traffic lights with the lanes belonging to the lanes and the use of the method in a vehicle. BACKGROUND

[0002] In autonomous or automatic driving of a motor vehicle, the motor vehicle must be able to cope with complex traffic light situations at a crossroads. For this purpose, an automatic recognition of the traffic lights present at the crossroads is required, wherein the recognition must also include an association of the traffic lights with the respective lanes for which the traffic lights are responsible. However, such an automatic recognition of the traffic lights at a crossroads can not only be used in autonomous driving, but also in assisted or semi-autonomous driving of a motor vehicle in order to assist the driver in driving the motor vehicle.

[0003] The automatic recognition of the traffic lights at a crossroads in autonomous driving pursues the following objectives:

[0004] - recommending a possible speed to the driver with respect to the traffic light installation;

[0005] - automatically stopping in front of a traffic light;

[0006] - alerting the driver against unintentional driving through a traffic light; and

[0007] - giving the driver a prompt to continue driving when the traffic light turns green again.

[0008] In order to achieve these examples, attempts are made to observe the traffic lights using a front camera in the vehicle. Here, the following measurement tasks must be solved:

[0009] - the camera recognizes the traffic light in advance and measures the traffic light in terms of position and orientation.

[0010] - the camera recognizes the lanes and the markings on the ground and thus determines the association of the traffic light with the lanes.

[0011] - the camera continuously observes the state of the relevant traffic light and passes this information on to the corresponding functions.

[0012] However, in reality, due to the large variety and complexity of crossroads situations, the association of the traffic light with the lanes is very difficult to determine, which leads to an excessively high error rate in associating the traffic light with the lanes.

[0013] The publication DE 10 2014 205 953 A1 relates to a method for analyzing the traffic situation of a vehicle with a navigation system and a sensor system, wherein the navigation system determines the vehicle position and the sensor system detects movement data of the vehicle and determines the movement behavior of the vehicle in the form of a movement trajectory of the vehicle, wherein the vehicle position and the movement trajectory of the vehicle enable the presence and / or type of at least one feature of the surrounding environment to be inferred, wherein a sensor of the sensor system detects this feature and stores at least one additional information about this feature.

[0014] The publication DE 10 2013 220 662 A1 describes a method for identifying a traffic situation when a vehicle is running on a road having a plurality of lanes, comprising the following features:

[0015] - optically detecting different surrounding data by means of a sensor device on the vehicle side, relating to the state of a signal facility, a road boundary, a lane boundary, at least one other marking on the road, and

[0016] - establishing a relationship between the state of the vehicle, which at least comprises a specification of the lane used by the vehicle, and the state of the signal facility valid for the state of the vehicle, in the case of using the surrounding data detected by means of the sensor device.

[0017] The later publication DE 10 2018 004 667 A1 relates to a method for determining a driving direction dependency of a light signal facility, wherein the light signal facility is detected by means of a camera system installed on a vehicle and the detected information is transmitted to an external computer unit and analyzed by means of the computer unit. It is provided here that the information detected by the camera system of the detected light signal facility is correlated with information stored in the external computer unit and the driving direction dependency is determined in dependence on the correlation.

[0018] The publication DE 10 2016 217 558 A1 relates to a method for associating a traffic light with a lane, wherein the recognized light sequence is compared with the expected light sequence. The publication also relates to an associated vehicle and an associated storage medium. SUMMARY

[0019] The task underlying the present invention is thus to provide a method for associating a traffic light with a lane and the use of the method in the case that the association of the traffic light with the lane is improved.

[0020] This task is solved by a method according to the invention and by the use of the method according to the invention. Preferred embodiments are the subject of the dependent claims.

[0021] The method according to the invention for associating traffic lights of a traffic light installation with lanes of a road section of a crossroad controlled by the traffic light installation, 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 vehicle platoon travel on the road section, has the following steps:

[0022] - identifying, by each vehicle of the vehicle platoon, a traffic light installation in front, said vehicle traveling on the road section and moving towards the traffic light installation;

[0023] - continuously detecting, by each vehicle, a traffic light state of a traffic light of the traffic light installation, a lane on which the vehicle is traveling, a vehicle speed and a vehicle position, wherein the detection starts at a predetermined distance from the traffic light installation and continues until the traffic light installation is passed;

[0024] - transmitting data recorded by each vehicle, i.e. the traffic light state, the lane on which the vehicle is traveling and the vehicle speed, in the form of a data set to a background computer; and

[0025] - performing, by the background computer, for each vehicle-specific data set, an analysis as follows:

[0026] - determining the identified traffic light installation on a map,

[0027] - identifying a movement of the vehicle through the road section,

[0028] - determining a traffic light state at the time when the vehicle passes the traffic light installation;

[0029] and

[0030] - creating, in the background computer, from the analyzed data sets of the vehicle platoon for the identified traffic light installation in the form of a matrix, an association table which can be called up by a vehicle, wherein the association table associates the traffic lights of the traffic light installation with the lanes of the road section, wherein

[0031] - in the association table, the traffic lights of the traffic light installation are listed with the traffic light states, wherein the traffic light states are limited to red and green, and

[0032] - In the association table, lanes of the road section are listed, wherein the lanes have binary values for going and for stopping as a function of the traffic light states.

[0033] Thus, the method according to the application determines the lane association of traffic lights using fleet data of a plurality of test vehicles. Here, the behavior of the driver is compared with the observation of the traffic light states in order to thus determine the correct association of traffic lights and lanes for each road section, in particular for a crossroads.

[0034] Preferably, the predetermined distance is 500 m, from which the detection of the traffic light states of the traffic light installation is started, wherein the predetermined distance can be shorter depending on the case, for example when there are a plurality of crossroads in succession within a short time.

[0035] Preferably, the association table of each traffic light installation is deposited in a database, whereby the association table can be called up for other vehicles.

[0036] In the association table, traffic lights of the traffic light installation are listed with the traffic light states, wherein the traffic light states are limited to red and green. By the said limitation to two traffic light states of interest, it is possible to pass through the traffic light or to have to stop, the traffic light states can be represented in the association table by one bit.

[0037] In the association table, lanes are also listed, wherein the lanes have binary values for going and for stopping as a function of the traffic light states.

[0038] Further preferably, the vehicles driving on the road section are manually controlled to create the data set. Preferably, the "fleet" of test vehicles for determining the association table is manually controlled in order to be able to call up the association of traffic lights and lanes for important crossroads for later autonomous driving in order to make autonomous driving safe.

[0039] In the application according to the application of the method according to the previously explained method for associating the traffic lights of a traffic light installation with the lanes of a road section controlled by the traffic light installation, the vehicle determines the position of the traffic light installation controlling the road section and requests the association table of the relevant traffic light installation from the database on the basis of the position, and the association table is used in an assistance system to assist the driver.

[0040] Preferably, the requested correlation table is analyzed in the vehicle together with current vehicle data, wherein the current vehicle data relate, inter alia, to the current lane backlog, the direction indicator setting and / or the navigation route. BRIEF DESCRIPTION OF DRAWINGS

[0041] Subsequently, a preferred embodiment of the application is explained with reference to the drawings. Herein:

[0042] Figure 1 A crossroad situation with corresponding traffic light installations is shown; whereas

[0043] Figure 2 A detail view of a traffic light situation for a crossroad with three lanes at the crossroad is shown. DETAILED DESCRIPTION

[0044] In Figure 1 a typical crossroad situation is described which is used to explain the method. The crossroad K is formed by the meeting of two carriageways Fb1 and Fb2. Herein, the first carriageway Fb1 comprises a total of six lanes F11 to F16, wherein three lanes extend upwards in Figure 1 Fb1 and the other three lanes F14 to F16 have an opposite direction. The second carriageway Fb2 extending perpendicular to the first carriageway Fb1 comprises two lanes F21 and F22 which extend in opposite directions. The traffic situation of the crossroad K with the two carriageways Fb1 and Fb2 is regulated by four traffic light installations A11, A12, A21 and A22 which are associated with each other, wherein the traffic light installation A11 is responsible for the lanes F11 to F13, the traffic light installation A12 is responsible for the lanes F14 to F16, the traffic light installation A21 is responsible for the lane F21 and the traffic light installation A22 is responsible for the lane F22. Furthermore, for the sake of completeness, a pedestrian crossing in the form of a zebra crossing Z is also drawn in Figure 1

[0045] Depending on the crossroad type, the traffic light installations A11, A12, A21 and A22 which are only symbolically drawn in Figure 1 may comprise a plurality of traffic lights with a special meaning and, for example, can be a function of the number of lanes. Thus, the traffic light installations A11 and A12 of the carriageway FB which has three lanes F11 to F16 in each direction regulate the driving behavior for going straight, for turning left and for turning right, so that the traffic light installations A11 and A12 are usually composed of a plurality of traffic lights which are responsible for the respective driving behavior. In other words, the mentioned traffic light installations A11 and A12 comprise separate traffic lights for going straight, for turning left and for turning right.

[0046] This is​Figure 2 As shown in the figure, the Figure 2 The diagram details the traffic situation at intersection K, specifically regarding traffic light facility A11 with zebra crossing Z and the lanes F11, F12, and F13 controlled by it. The right-hand portion (AR) and left-hand portion (AL) of traffic light facility A11 control traffic in lanes F11, F12, and F13. Specifically, the right-hand portion (AR) uses traffic light AR1 to regulate straight-ahead traffic in lanes F11, F12, and F13, and traffic light AR2 to regulate right-turning traffic in lane F11. Similarly, the left-hand portion (AL) uses traffic light AL1 to regulate straight-ahead traffic in lanes F11, F12, and F13, and traffic light AL2 to regulate left-turning traffic in lane F13.

[0047] The method includes the following steps:

[0048] a) Vehicles pre-identify traffic signal facilities and the status of each traffic light—green, yellow, red, red / yellow—and identify which lane the vehicle is in. This data is continuously recorded until the vehicle has passed the traffic light. Speed ​​is also recorded to determine whether the vehicle has stopped in front of the traffic light or may proceed without stopping. In real-world scenarios, there is a possibility that the traffic light status may not be correctly identified. This could be due to poor visibility conditions (glare, pollution, etc.), obstruction of the traffic light (e.g., a truck (LkW) within sight), or the traffic light being outside the camera's visible range or angle. In these cases, only the identified traffic light status is reported; the vehicle is unaware of missing statuses and therefore cannot communicate the missing statuses.

[0049] b) Send the recorded data to the backend system.

[0050] c) Perform the following steps in the background:

[0051] i) Locate the identified traffic signal facilities on the map.

[0052] ii) Identify the movement of vehicles as they pass through the intersection.

[0053] iii) Determine the status of traffic lights when crossing the intersection.

[0054] For in Figure 2The traffic light installation A11 shown in Fig. 1 is explained in the following table with exemplary data sets as they are transmitted to the background computer or present in the background computer for processing:

[0055] Table 1:

[0056] Right turn traffic light AR2 Traffic light without arrow AR1 and AR2 Left turn traffic light AL2 Lane on which the vehicle is driving Go (1) / Stop (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

[0057] Here, the following applies for the driving behavior at the traffic light:

[0058] Passing = "1", stopping = "0".

[0059] The data sets transmitted to the background computer thus contain information about the traffic light status, the lane occupied by the vehicle and the behavior at the intersection, i.e. whether the vehicle is passing or stopping. For example, the first line means: the vehicle on the left lane F13 is passing, wherein the traffic light for left turns AL2 is "green" and all other traffic lights show "red". From this it can be concluded that the traffic light AL2 is associated with the left lane F13.

[0060] e) If the starting point is that the intersection is only driven through when green, then the association with the possible lanes can now be achieved. From the example above it can be concluded that when the left traffic light AL2 is green, driving on lane 3 is allowed. Driving on lane 2 is not allowed, however. Information about lane 1 does not yet exist. In addition, when the traffic lights without arrows AR1, AL1 are green, driving on all lanes is allowed.

[0061] f) Depending on the traffic light system, the traffic light display can be interpreted in several ways. For example, in a traffic light installation, not only straight ahead but also left is green. However, the vehicle only chooses to drive left. This does not lead to a functional impairment, however, since this is solved by an association table which must be detected for each traffic light installation. This detection is achieved by several passages by the "vehicle fleet". Here, in the corresponding state of the traffic light installation, each vehicle determines the state of the lane in the form of a Boolean variable with the values "0" or "1".

[0062] In this case, possible traffic lights which are not recognized by the vehicle and thus not reported are not considered for the expansion of the table and do not contribute to the information build-up, but also do not lead to an error state and can be tolerated in this respect.

[0063] According to the data transmitted to the background by the "platoon" shown in table 1, an association table is formulated in matrix form for the crossroad section involving the traffic light installation Al l, and this association example is reproduced in table 2, for example. The association table for the other crossroad sections involving the traffic light installations A12, A21 and A22 shown in table 1 looks like this, correspondingly. Figure 1

[0064] Table 2:

[0065] Right traffic light AR2 (1 = green, 0 = red) Traffic light without arrow AR1, AL1 (1 = green, 0 = red) Left traffic light AL2 (1 = green, 0 = red) Lane F11 (0 = driving, 1 = stopped) Lane F12 (0 = driving, 1 = stopped) Lane F13 (0 = driving, 1 = stopped) 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

[0066] g) If the association is determined, the association is deposited in the database in the form of an association table for each traffic light installation and is transmitted to the vehicle, for example as a layer of the map.

[0067] h) If the vehicle now drives again to the traffic light installation, the possible free lanes can be determined from the association table.

[0068] i) This situation is then analyzed by the functions together with the current lane backlog, the turn signal setting, the navigation route, etc.

[0069] Further improvements are possible if the entries in the table can be determined directly in the vehicle. In this case, although the logic circuitry in the vehicle needs to be expanded, the data volume can be significantly reduced in this way, since here only the lane-related status of the traffic light status needs to be transmitted.

[0070] List of reference signs

[0071] K crossroad

[0072] Fbl lane 1 of lane 1

[0073] Fb2 lane 2 of lane 1

[0074] Fl 1 lane 1 of lane 1

[0075] Fl 2 lane 2 of lane 1

[0076] Fl 3 lane 3 of lane 1

[0077] Fl 4 lane 4 of lane 1

[0078] Fl 5 lane 5 of lane 1

[0079] Fl 6 lane 6 of lane 1

[0080] F21 lane 1 of lane 2 ​

[0081] F22 lane 2 of carriageway 2

[0082] A11 traffic light installation for lanes F11 to F13

[0083] A12 traffic light installation for lanes F14 to F16

[0084] A21 traffic light installation for lane F21

[0085] A22 traffic light installation for lane F22

[0086] Z zebra crossing

[0087] AR right-hand part of traffic light installation

[0088] AR1 straight-ahead traffic light

[0089] AR2 right-turn traffic light

[0090] AL left-hand part of traffic light installation

[0091] AL1 straight-ahead traffic light

[0092] AL2 left-turn traffic light

Claims

1. Method for associating traffic lights (AR1, AR2, AL1, AL2) of a traffic light installation (A11) controlling a traffic road section (Fb1) of a crossroad (K) to lanes (F11, F12, F13) of the traffic road section (Fb1), wherein the traffic road section (Fb1) has at least one lane (F11, F12, F13) with a common driving direction and a plurality of vehicles forming a vehicle fleet are driving on the traffic road section (Fb1), with the following steps: - identifying by each vehicle of the vehicle fleet a traffic light installation (A11) in front, which the vehicle is driving on the traffic road section (Fb1) and is moving towards the traffic light installation (A11), - continuously detecting by each vehicle a traffic light state of the traffic lights (AR1, AR2, AL1, AL2) of the traffic light installation (A11), a lane (F11, F12, F13) on which the vehicle is driving, a vehicle speed and a vehicle position, wherein the detection starts at a predetermined distance from the traffic light installation (A11) and continues until the traffic light installation (A11) is passed, - transmitting the data recorded by each vehicle, i.e. the traffic light state, the lane on which the vehicle is driving and the vehicle speed, in the form of a data set to a background computer, - performing by the background computer for each vehicle-specific data set of the vehicles of the vehicle fleet the following analysis: - determining the identified traffic light installation (A11) on a map, - identifying the movement of the vehicle through the traffic road section (Fb1), - determining the traffic light state of the traffic lights (AR1, AR2, AL1, AL2) at the time when the vehicle passes the traffic light installation (A11), - summarizing the analysis of the data of each vehicle of the vehicle fleet into a corresponding data set related to the carriageway section, wherein the data set comprises the traffic light states of the traffic lights (AR1, AR2, AL1, AL2) of the carriageway section limited to red and green, the lane of the carriageway section driven by the corresponding vehicle and the driving behavior of the vehicle at the traffic lights (AR1, AR2, AL1, AL2), i.e. whether the vehicle remains stopped or proceeds, and - creating in the background computer from the analyzed data sets of the vehicle fleet for the identified traffic light installation (A11) in the form of a matrix an association table which can be called up by a vehicle, wherein the association table associates the traffic lights (AR1, AR2, AL1, AL2) of the traffic light installation (A11) to the lanes (F11, F12, F13) of the traffic road section (Fb1). - in the association table, listing the traffic lights (AR1, AR2, AL1, AL2) of the traffic light installation (A11) having the traffic light states, wherein the traffic light states are limited to red and green, and - in the association table, listing the lanes (F11, F12, F13) of the carriageway section (Fb1), wherein the lanes (F11, F12, F13) have a binary value for passing and for stopping as a function of the traffic light states.

2. The method of claim 1, wherein, The predetermined distance is 500 m or less.

3. The method according to claim 1 or 2, characterized in that, The association table of each traffic light installation (A11) is deposited in a database.

4. The method according to claim 1 or 2, characterized in that, The data sets are created manually by controlling a vehicle driving on the carriageway section (Fb1).

5. Use of the method according to any one of the preceding claims for associating traffic lights of a traffic light installation (A11) controlling a carriageway section (Fb1) with lanes (F11, F12, F13) of the carriageway section (Fb1) in a vehicle driving on the carriageway section (Fb1), wherein the vehicle determines the position of the traffic light installation (A11) controlling the carriageway section (Fb1) and requests the association table of the relevant traffic light installation (A11) from a database on the basis of the position, and uses the association table in an assistance system to assist the driver.

6. Use of the method according to claim 5, characterized in that The requested association table is analyzed in the vehicle together with current vehicle data.

7. Use of the method according to claim 5 or 6, characterized in that The current vehicle data relate to the current lane backlog, the indicator setting and / or the navigation route. The current vehicle data relate to the current lane backlog, the indicator setting and / or the navigation route.

Images