Method and apparatus for autonomous driving control, electronic device, and storage medium

1. A method for autonomous driving control, comprising:

obtaining, by a processor, movement data of a pedestrian; wherein the movement data comprises: a velocity component of the pedestrian along a width direction of a lane and a time of duration that the pedestrian cuts into a driving path of an autonomous vehicle from one side, wherein, the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from one side indicates a likelihood that the pedestrian will intrude into the driving path of the autonomous vehicle;

determining, by the processor, a movement direction of the pedestrian according to the movement data and movement information of the pedestrian;

generating, by the processor, a driving strategy for the autonomous vehicle according to the movement direction of the pedestrian; and

controlling, by the processor, driving of the autonomous vehicle according to the driving strategy;

wherein the obtaining, by the processor, movement data of the pedestrian comprises:

obtaining, by the processor, a position of the lane where the pedestrian is located and the movement velocity of the pedestrian;

resolving, by the processor, the movement velocity of the pedestrian in accordance to an extension direction and the width direction of the lane to obtain a velocity component along the extension direction of the lane and the velocity component along the width direction of the lane; and

determining, by the processor, the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from the one side according to the velocity component of the pedestrian along the width direction of the lane and the position of the lane where the pedestrian is located;

wherein the time of duration refers to a time of duration of keeping the movement velocity for cutting into the driving path of the autonomous vehicle from one side non-zero.

2. The method according to claim 1 , wherein the obtaining, by the processor, the position of the lane where the pedestrian is located comprises:

capturing, by the processor, an environmental image of surroundings of the autonomous vehicle;

extracting, by the processor, lane information from the environmental image, wherein the lane information comprises: positions of lane distribution, a lane width, and a lane direction; and

determining, by the processor, the position of the lane where the pedestrian is located according to the lane information.

3. The method according to claim 1 , wherein the determining, by the processor, the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from the one side according to the velocity component of the pedestrian along the width direction of the lane and the position of the lane where the pedestrian is located comprises:

obtaining, by the processor, velocity components of the pedestrian along the width direction of the lane corresponding to multiple moments within a preset time period;

performing, by the processor, a smoothing processing on the velocity components of the pedestrian along the width direction of the lane corresponding to the multiple moments to obtain a smoothing processed velocity component; and

determining, by the processor, the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from the one side according to the smoothing processed velocity component and the position of the lane where the pedestrian is located.

4. The method according to claim 3 , wherein the movement information of the pedestrian comprises: an average velocity of the pedestrian in the time of duration, a distance between the pedestrian and the autonomous vehicle, and a standard deviation of the movement velocity of the pedestrian.

5. The method according to claim 1 , wherein the determining, by the processor, the movement direction of the pedestrian according to the movement data and movement information of the pedestrian comprises:

constructing, by the processor, a decision tree model, wherein the decision tree model comprises: a gradient boosted decision tree (GBDT) model; and

inputting, by the processor, the movement data and the movement information of the pedestrian into the decision tree model, wherein the decision tree model outputs the movement direction of the pedestrian; wherein the movement direction of the pedestrian comprises: an extension direction of the lane, a leftward direction for cutting into the lane and a rightward direction for cutting into the lane.

6. The method according to claim 1 , wherein the generating, by the processor, the driving strategy for the autonomous vehicle according to the movement direction of the pedestrian comprises:

generating, by the processor, the driving strategy for the autonomous vehicle according to the movement direction of the pedestrian, wherein the driving strategy comprises: adjustment of a driving velocity and adjustment of a driving direction.

7. The method according to claim 1 , wherein the method further comprises:

determining, by the processor, an avoidance strategy for the autonomous vehicle according to the movement direction of the pedestrian, driving information of the autonomous vehicle and the movement information of the pedestrian; wherein the driving information of the autonomous vehicle comprises: a driving velocity of the autonomous vehicle, information about a position of the lane where the autonomous vehicle is located, and a distance between the autonomous vehicle and the pedestrian; and

controlling, by the processor, driving of the autonomous vehicle according to the avoidance strategy.

8. An electronic device, comprising:

at least one processor; and

a memory connected in communication with the at least one processor; wherein

the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

obtain movement data of a pedestrian; wherein the movement data comprises: a velocity component of the pedestrian along a width direction of a lane and a time of duration that the pedestrian cuts into a driving path of an autonomous vehicle from one side, wherein, the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from one side indicates a likelihood that the pedestrian will intrude into the driving path of the autonomous vehicle;

determine a movement direction of the pedestrian according to the movement data and movement information of the pedestrian;

generate a driving strategy for the autonomous vehicle according to the movement direction of the pedestrian; and

control driving of the autonomous vehicle according to the driving strategy;

wherein the at least one processor is further enabled to:

obtain a position of the lane where the pedestrian is located and the movement velocity of the pedestrian;

resolve the movement velocity of the pedestrian in accordance to an extension direction and the width direction of the lane to obtain a velocity component along the extension direction of the lane and the velocity component along the width direction of the lane; and

determine the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from the one side according to the velocity component of the pedestrian along the width direction of the lane and the position of the lane where the pedestrian is located;

wherein the time of duration refers to a time of duration of keeping the movement velocity for cutting into the driving path of the autonomous vehicle from one side non-zero.

9. The device according to claim 8 , wherein the at least one processor is further enabled to:

capture an environmental image of surroundings of the autonomous vehicle;

extract lane information from the environmental image, wherein the lane information comprises: positions of lane distribution, a lane width, and a lane direction; and

determine the position of the lane where the pedestrian is located according to the lane information.

10. The device according to claim 8 , wherein the at least one processor is further enabled to:

obtain velocity components of the pedestrian along the width direction of the lane corresponding to multiple moments within a preset time period;

perform a smoothing processing on the velocity components of the pedestrian along the width direction of the lane corresponding to the multiple moments to obtain a smoothing processed velocity component; and

determine the time of duration that the pedestrian cuts into the driving path of the autonomous vehicle from the one side according to the smoothing processed velocity component and the position of the lane where the pedestrian is located.

11. The device according to claim 10 , wherein the movement information of the pedestrian comprises: an average velocity of the pedestrian in the time of duration, a distance between the pedestrian and the autonomous vehicle, and a standard deviation of the movement velocity of the pedestrian.

12. The device according to claim 8 , wherein the at least one processor is further enabled to:

construct a decision tree model, wherein the decision tree model comprises: a gradient boosted decision tree (GBDT) model; and

input the movement data and the movement information of the pedestrian into the decision tree model, wherein the decision tree model outputs the movement direction of the pedestrian; wherein the movement direction of the pedestrian comprises: an extension direction of the lane, a leftward direction for cutting into the lane and a rightward direction for cutting into the lane.

13. The device according to claim 8 , wherein the at least one processor is further enabled to generate the driving strategy for the autonomous vehicle according to the movement direction of the pedestrian, wherein the driving strategy comprises: adjustment of a driving velocity and adjustment of a driving direction.

14. A non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to claim 1 .