Method for automatically producing and updating a data set for an autonomous vehicle
A method for the automatic production and updating of a data set for an autonomous vehicle, in which at least one traffic light and a switching state of the at least one traffic light are registered; at least one road marking is ascertained; a trajectory of at least one vehicle traveling ahead is registered; and the collected data are used for producing and updating a data set, and based on the at least one detected trajectory, the at least one switching state of the at least one traffic light and the at least one ascertained road marking, at least one traffic lane is allocated to at least one traffic light. In addition, an autonomous or partially autonomous vehicle is described for carrying out the method.
1. A method for an autonomous vehicle, the method comprising:
the autonomous vehicle using a sensor system of the autonomous vehicle to obtain sensor signals;
based on the sensor signals, a processor of the autonomous vehicle:
identifying at least one traffic light and a respective switching state of each of the identified at least one traffic light;
ascertaining at least one road marking; and
detecting a respective trajectory of at least one vehicle traveling spatially ahead of the autonomous vehicle, wherein, based on the detected trajectory of the at least one vehicle traveling spatially ahead, the respective switching state of the each of the identified at least one traffic light, and the at least one ascertained road marking, at least one traffic lane k identified as being allocated to the identified at least one traffic light; and
the processor of the autonomous vehicle performing an autonomous control of the autonomous vehicle according to a determination of which light signals of the at least one traffic light apply to the autonomous vehicle based on the identified allocation of the at least one traffic lane to the identified at least one traffic light.
2. The method of claim 1 , wherein the allocation is performed by an external server unit based on data that, at different points in time, are obtained by different vehicles and transmitted to the external server unit for evaluation.
3. The method of claim 1 , wherein a switching cycle of the at least one traffic light is detected through the measurements obtained by different vehicles at different points in time.
4. The method of claim 1 , wherein a duration of at least one switching state of the at least one traffic light is ascertained using at least two vehicle-based measurements.
5. The method of claim 1 , wherein switching states of at least two traffic light systems relative to one another along a driving route are detected and stored.
6. The method of claim 1 , wherein the ascertained data from at least two vehicles are collected on an external server unit and processed to form a model of traffic light systems and their mutual dependencies.
7. The method of claim 6 , wherein the model and the collected data are automatically stored in a vehicle-external highly automated driving (HAD) map or are used to correct a vehicle-external HAD map.
8. The method of claim 1 , wherein the takes place in the autonomous vehicle.
9. An autonomous or partially autonomous vehicle, comprising: at least one sensor; and
a control unit, wherein the control unit is configured to:
use the at least one sensor to obtain sensor signals;
based on the sensor signals: identify at least one traffic light and a respective switching state of each of the identified at least one traffic light;
ascertain at least one road marking; and
detect a respective trajectory of at least one vehicle traveling spatially ahead of the autonomous vehicle, wherein, based on the detected trajectory of the at least one vehicle traveling spatially ahead, the respective switching state of the each of the identified at least one traffic light, and the at least one ascertained road marking, at least one traffic lane is identified as being allocated to the identified at least one traffic light; and
perform an autonomous control of the autonomous vehicle according to a determination of which light signals of the at least one traffic light apply to the autonomous vehicle based on the identified allocation of the at least one traffic lane to the identified at least one traffic light.
10. The vehicle of claim 9 , wherein a vehicle-external highly automated driving (HAD) map is use-able for the autonomous vehicle for navigation purposes, the HAD map being updated based on the allocation.
11. The vehicle of claim 10 , wherein traffic light positions and ascertained data sets pertaining to traffic lights are projectable from the vehicle-external highly automated driving (HAD) map into a vehicle-internal image-coordinate system of at least one vehicle-internal camera.