Granted Patent B2

US 12,397,826 · App. 17/871,522 · Granted Aug 26, 2025

Inventor: Peng Wu (Beijing, CN)

Assignee: XIAOMI EV TECHNOLOGY CO., LTD.

B60W60/0016B60W40/04G06T7/246G06V20/58G08G1/166B60W2420/403G06T2207/30241G06T2207/30256G06T2207/30261

Abstract

Claims (90)

1. A method for detecting a drivable area, performed by an electronic device, comprising:

determining a value of N based on a width of a road where a vehicle is located;

after determining the value of N, collecting N consecutive video frames of the road when the vehicle is driving, where N is a positive integer greater than 1;

determining a historical trajectory of a dynamic obstacle and position information of a static obstacle included in the N consecutive video frames by analyzing the N consecutive video frames with a 3D detection algorithm;

correcting the historical trajectory and the position information based on a preset rule;

determining a predicted trajectory of the dynamic obstacle based on the corrected historical trajectory;

determining the drivable area of the vehicle based on the predicted trajectory and the corrected position information;

performing automatic driving based on the determined drivable area; and

determining a type of the dynamic obstacle, and correcting the historical trajectory based on a correction parameter corresponding to the type.

2. The method of claim 1 , wherein correcting the historical trajectory and the position information based on the preset rule comprises:

determining positioning information of the vehicle when collecting each of the video frames;

obtaining map information corresponding to the positioning information; and

correcting the historical trajectory and the position information based on the map information.

3. The method of claim 2 , wherein a road trajectory comprises a plurality of key points, correcting the historical trajectory and the position information based on the map information comprises:

determining information of each key point for describing a road trajectory based on the map information;

determining a comparison result by comparing the historical trajectory and the position information to the information of each key point; and

correcting the historical trajectory and the position information based on the comparison result.

4. The method of claim 3 , wherein correcting the historical trajectory and the position information based on the comparison result comprises one of:

in response to a deviation between information of any point in the historical trajectory or position information and corresponding key point information being less than a set threshold, stopping correction of the historical trajectory or the position information; and

in response to a deviation between information of any point in the historical trajectory or position information and corresponding key point information being larger than a set threshold, replacing the information of any point in the historical trajectory or position information with the corresponding key point information.

5. The method of claim 1 , wherein determining the predicted trajectory of the dynamic obstacle comprises:

obtaining map coding information by encoding map information corresponding to a road lane;

obtaining a plurality of prediction positions by inputting the map coding information and the historical trajectory into a trajectory prediction model trained in advance; and

determining the predicted trajectory of the dynamic obstacle by splicing the plurality of prediction positions in order.

6. The method of claim 1 , wherein determining the historical trajectory of the dynamic obstacle and position information of the static obstacle comprises:

determining coordinate information of the vehicle when collecting each of the video frames;

obtaining a position of the static obstacle and a position of the dynamic obstacle in a real world at each moment corresponding to each of the video frames, by converting the static obstacle and the dynamic obstacle into a world coordinate system according to the positioning information of the vehicle and a 3D detection result; and

obtaining the historical trajectory of the dynamic obstacle by splicing the positions of the dynamic obstacle in the world coordinate system, and obtaining the position information of the static obstacle by splicing the positions of the static obstacle in the world coordinate system.

7. The method of claim 1 , wherein determining the value of N further comprises:

determining a current driving speed of the vehicle;

determining a speed interval corresponding to the current driving speed; and

determining the value of N matching the speed interval, wherein a correspondence between speed intervals and values of N is preset in the vehicle.

8. A mobile device, comprising:

a processor coupled; and

a memory storing non-transitory computer-readable instructions that, when executed by the processor, control the processor to

determine a value of N based on a width of a road where the vehicle is located;

after determining the value of N, receive N consecutive video frames of the road collected when a vehicle is driving, where N is a positive integer greater than 1;

determine a historical trajectory of a dynamic obstacle and position information of a static obstacle included in the N consecutive video frames by analyzing the N consecutive video frames with a 3D detection algorithm;

correct the historical trajectory and the position information based on a preset rule;

determine a predicted trajectory of the dynamic obstacle based on the corrected historical trajectory;

determine a drivable area of the vehicle based on the predicted trajectory and the corrected position information; and

perform automatic driving based on the determined drivable area;

wherein the processor is further configured to: determine a type of the dynamic obstacle, and correct the historical trajectory based on a correction parameter corresponding to the type.

9. The mobile device of claim 8 , wherein the processor is further configured to:

determine positioning information of the vehicle when collecting each of the video frames;

obtain map information corresponding to the positioning information; and

correct the historical trajectory and the position information based on the map information.

10. The mobile device of claim 9 , wherein a road trajectory comprises a plurality of key points, the processor is further configured to:

determine information of each key point for describing a road trajectory based on the map information;

determine a comparison result by comparing the historical trajectory and the position information to the information of each key point; and

correct the historical trajectory and the position information based on the comparison result.

11. The mobile device of claim 10 , wherein the processor is further configured to perform one of:

in response to a deviation between information of any point in the historical trajectory or position information and corresponding key point information being less than a set threshold, stopping correction of the historical trajectory or the position information; and

in response to a deviation between information of any point in the historical trajectory or position information and corresponding key point information being larger than a set threshold, replacing the information of any point in the historical trajectory or position information with the corresponding key point information.

12. The mobile device of claim 8 , wherein the processor is further configured to:

obtain map coding information by encoding map information corresponding to a road lane;

obtain a plurality of prediction positions by inputting the map coding information and the historical trajectory into a trajectory prediction model trained in advance; and

determine the predicted trajectory of the dynamic obstacle by splicing the plurality of prediction positions in order.

13. The mobile device of claim 8 , wherein the processor is further configured to:

determine coordinate information of the vehicle when collecting each of the video frames;

obtain a position of the static obstacle and a position of the dynamic obstacle in a real world at each moment corresponding to each of the video frames, by converting the static obstacle and the dynamic obstacle into a world coordinate system according to the positioning information of the vehicle and a 3D detection result; and

obtain the historical trajectory of the dynamic obstacle by splicing the positions of the dynamic obstacle in the world coordinate system, and obtain the position information of the static obstacle by splicing the positions of the static obstacle in the world coordinate system.

14. The mobile device of claim 8 , wherein the processor is further configured to:

determine a current driving speed of the vehicle;

determine a speed interval corresponding to the current driving speed; and

determine the value of N matching the speed interval, wherein a correspondence between speed intervals and values of N is preset in the vehicle.

15. A non-transitory computer readable storage medium having instructions stored thereon, wherein when the instructions are executed by a processor of an electronic device, the electronic device is caused to execute a method for detecting a drivable area, the method comprising:

determining a value of N based on a width of a road where a vehicle is located;

after determining the value of N, collecting N consecutive video frames of the road when the vehicle is driving, where N is a positive integer greater than 1;

determining a historical trajectory of a dynamic obstacle and position information of a static obstacle included in the N consecutive video frames by analyzing the N consecutive video frames with a 3D detection algorithm;

correcting the historical trajectory and the position information based on a preset rule;

determining a predicted trajectory of the dynamic obstacle based on the corrected historical trajectory;

determining the drivable area of the vehicle based on the predicted trajectory and the corrected position information; and

performing automatic driving based on the determined drivable area;

wherein the method further comprises: determining a type of the dynamic obstacle, and correcting the historical trajectory based on a correction parameter corresponding to the type.

16. The storage medium of claim 15 , wherein correcting the historical trajectory and the position information based on the preset rule comprises:

determining positioning information of the vehicle when collecting each of the video frames;

obtaining map information corresponding to the positioning information; and

correcting the historical trajectory and the position information based on the map information.

17. The storage medium of claim 16 , wherein a road trajectory comprises a plurality of key points, correcting the historical trajectory and the position information based on the map information comprises:

determining information of each key point for describing a road trajectory based on the map information;

determining a comparison result by comparing the historical trajectory and the position information to the information of each key point; and

correcting the historical trajectory and the position information based on the comparison result.

18. The storage medium of claim 17 , wherein correcting the historical trajectory and the position information based on the comparison result comprises one of:

in response to a deviation between information of any point in the historical trajectory or position information and corresponding key point information being less than a set threshold, stopping correction of the historical trajectory or the position information; and

in response to a deviation between information of any point in the historical trajectory or position information and corresponding key point information being larger than a set threshold, replacing the information of any point in the historical trajectory or position information with the corresponding key point information.

19. The storage medium of claim 15 , wherein determining the predicted trajectory of the dynamic obstacle comprises:

obtaining map coding information by encoding map information corresponding to a road lane;

obtaining a plurality of prediction positions by inputting the map coding information and the historical trajectory into a trajectory prediction model trained in advance; and

determining the predicted trajectory of the dynamic obstacle by splicing the plurality of prediction positions in order.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (28)