Computer-implemented method and system for classifying a traffic situation

1 . A computer-implemented method for classifying a predefined traffic situation comprised by a data record of environment data of a motor vehicle, the method comprising:

providing a first data record or a data stream of sensor data comprising a range of traffic situations from a journey of an ego vehicle and/or a fellow vehicle collected by at least one on-board environment perception sensor;

applying a directed graph to the first data record, wherein nodes of the directed graph segment the first data record into at least one segment of a movement behavior of the ego vehicle and/or the fellow vehicle relative to a vehicle environment according to a first condition met in a time interval, and wherein edges of the directed graph symbolize links between the respective nodes;

classifying the predefined traffic situation, when all of the specified segments meet a second condition of the predefined traffic situation; and

outputting a class representing the predefined traffic situation and/or a respective start and end time of a second data record comprising the segment representing the predefined traffic situation.

2 . The computer-implemented method according to claim 1 , wherein the predefined traffic situation to be classified is determined in advance, and wherein the directed graph determines segments of the predefined traffic situation in each case.

3 . The computer-implemented method according to claim 1 , wherein each node has at least one input and exactly one output, and wherein at least one input of each node is given by an edge of another node or by the first data record.

4 . The computer-implemented method according to claim 1 , wherein each node has an algorithm that is applied to input data of the node, and wherein the algorithm classifies whether the first condition of the movement behavior of the ego vehicle and/or the fellow vehicle relative to the vehicle environment is met.

5 . The computer-implemented method according to claim 4 , wherein the algorithm of the respective node outputs a respective start and end time of the segment representing the movement behavior when the first condition is met.

6 . The computer-implemented method according to claim 1 , wherein the segment of the movement behavior of the ego vehicle and/or the fellow vehicle relative to the vehicle environment is a time interval comprising a start and end time of the movement behavior of the ego vehicle and/or the fellow vehicle comprised by the first data record or the data stream, relative to the vehicle environment.

7 . The computer-implemented method according to claim 6 , wherein the directed graph comprises a first layer that comprises at least one node, an input of which is the first data record, wherein a second layer of the directed graph comprising at least one first node and a second node segments the first data record in each case corresponding to the first condition into at least one segment of the movement behavior of the ego vehicle and/or the fellow vehicle relative to the vehicle environment, and wherein a third layer of the directed graph comprising at least one node classifies the predefined traffic situation using time intervals output by the nodes of the second layer, when a combination of the specified time intervals satisfies the second condition of the predefined traffic situation.

8 . The computer-implemented method according to claim 7 , wherein the second condition of the predefined traffic situation specifies that the segments of the ego vehicle and/or at least one segment of a movement behavior of the fellow vehicle determined according to the first condition take place in a predefined sequence and/or within a predefined time interval.

9 . The computer-implemented method according to claim 1 , wherein the movement behavior of the ego vehicle and/or the fellow vehicle relative to the vehicle environment includes all vehicle actions representing the movement behavior detectable in the first data record.

10 . The computer-implemented method according to claim 1 , wherein the movement behavior of the ego vehicle and/or the fellow vehicle relative to the vehicle environment is a lateral and or longitudinal behavior of the ego vehicle relative to a traffic infrastructure and/or at least one fellow vehicle and/or the lateral and/or longitudinal behavior of the fellow vehicle is relative to a traffic infrastructure and/or at least one ego vehicle, wherein the movement behavior of the ego vehicle and/or the fellow vehicle relative to the vehicle environment includes lane keeping, lane change, turning, constant or changing acceleration and resulting speed, flashing brake lights, passing an object in the vehicle environment and/or recognizing a traffic sign.

11 . The computer-implemented method according to claim 1 , wherein the sensor data of the journey of the ego vehicle and/or the fellow vehicle collected by at least one on-board environment perception sensor are position data of a GNSS sensor, IMU data, camera data, LiDAR data, radar data and/or ultrasonic data.

12 . The computer-implemented method according to claim 1 , wherein a virtual test is performed on at basis of an output second data record for a validation of an automated driving function of the motor vehicle.

13 . The computer-implemented method according to claim 1 , wherein the output of the directed graph and an output of another directed graph is used to classify another predefined traffic situation, and wherein when a combination of the outputs of the directed graph and a further directed graph satisfy a third condition of a further predefined traffic situation, the further predefined traffic situation is classified.

14 . A system for classifying a predefined traffic situation comprised by a data record of environment data of a motor vehicle, the system comprising a processor and memory configured to:

provide a first data record or a data stream of sensor data comprising a plurality of traffic situations of a journey of an ego vehicle and/or a fellow vehicle collected by at least one on-board environment perception sensor;

apply a directed graph to the first data record, wherein nodes of the directed graph segment the first data record in each case into at least one segment of a movement behavior of the ego vehicle and/or the fellow vehicle relative to a vehicle environment according to a first condition satisfied in a time interval, wherein edges of the directed graph symbolize links between the respective nodes, the processor or a second processor classifying the predefined traffic situation when all of the specified segments meet a second condition of the predefined traffic situation; and

output a second data record comprising a class representing the predefined traffic situation and/or a respective start and end time of the segment representing the predefined traffic situation.