IP Library › Granted Patent US 12,202,473
Granted Patent B2
US 12,202,473 · App. 17/669,355 · Granted Jan 21, 2025

Obstacle tracking method, storage medium and unmanned driving device

Inventors: Huaxia Xia (Beijing, CN); Shanbo Cai (Beijing, CN)
Assignee: Beijing Sankuai Online Technology Co., Ltd.
B60W30/0956B60W40/04B60W2420/408B60W2554/4041G06T2207/30261
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,202,473
App. No.
17/669,355
Granted
Jan 21, 2025
Kind
B2
Abstract

An obstacle tracking method and apparatus, a storage medium and an unmanned driving device are provided. An obstacle aggregation region in a first point cloud may be determined according to position information of each obstacle in the first point cloud acquired by the unmanned driving device. Then, an aggregated obstacle and a non-aggregated obstacle in the second point cloud acquired by the unmanned driving device are determined according to the obstacle aggregation region. In addition, a matching result of each aggregated obstacle is respectively determined based on a group matching rule, and a matching result of each non-aggregated obstacle is respectively determined based on a non-group matching rule. Finally, an obstacle tracking result is determined according to the matching result of each obstacle in the second point cloud.

Claims (100)

1. An obstacle tracking method, comprising:

obtaining two frames of laser point clouds acquired by an unmanned driving device as a first point cloud and a second point cloud;

determining an obstacle aggregation region in the first point cloud according to position information of each obstacle in the first point cloud;

determining, according to a position of the obstacle aggregation region, at least one obstacle in the second point cloud within the obstacle aggregation region as at least one aggregated obstacle, and at least one another obstacle, except the at least one aggregated obstacle, in the second point cloud as at least one non-aggregated obstacle; and

for each of the at least one aggregated obstacle,

determining matching parameter values between the aggregated obstacle and each obstacle in the first point cloud by using a group matching rule, and

determining a matching result of the aggregated obstacle according to the matching parameter values between the aggregated obstacle and each obstacle in the first point cloud;

for each of the at least one non-aggregated obstacle,

determining matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using a non-group matching rule, and

determining a matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud, wherein a constraint condition of the group matching rule is stronger than a constraint condition of the non-group matching rule; and

determining an obstacle tracking result according to the matching result of each of the at least one aggregated obstacle and the matching result of each of the at least one non-aggregated obstacle.

2. The method according to claim 1 , wherein determining the obstacle aggregation region in the first point cloud according to the position information of each obstacle in the first point cloud comprises:

determining a distance between every two obstacles in the first point cloud according to the position information of each obstacle in the first point cloud; and

determining the obstacle aggregation region in the first point cloud according to the distance between every two obstacles in the first point cloud.

3. The method according to claim 1 , wherein determining the matching parameter values between the aggregated obstacle and each obstacle in the first point cloud by using the group matching rule comprises at least one of:

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a distance between the aggregated obstacle and the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a size of the aggregated obstacle and a size of the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a quantity of points comprised in the aggregated obstacle in the second point cloud and a quantity of points comprised in the obstacle in the first point cloud; or

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a type of the aggregated obstacle and a type of the obstacle in the first point cloud.

4. The method according to claim 1 , wherein determining the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using the non-group matching rule comprises at least one of:

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a distance between the non-aggregated obstacle and the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a size of the non-aggregated obstacle and a size of the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a quantity of points comprised in the non-aggregated obstacle in the second point cloud and a quantity of points comprised in the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a type of the non-aggregated obstacle and a type of the obstacle in the first point cloud; or

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to an orientation of the non-aggregated obstacle and an orientation of the obstacle in the first point cloud.

5. The method according to claim 1 , further comprising:

changing the aggregated obstacle to a non-aggregated obstacle in response to the matching result of the aggregated obstacle being not determined.

6. The method according to claim 1 , wherein determining the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using the non-group matching rule comprises:

determining obstacles in the first point cloud that do not match any aggregated obstacle in the second point cloud as a first obstacle set; and

determining matching parameter values between the non-aggregated obstacle and each of the obstacles in the first obstacle set by using the non-group matching rule; and, wherein

determining the matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud comprises:

determining a matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each of the obstacles in the first obstacle set.

7. The method according to claim 1 , further comprising:

determining that each obstacle comprised in the second point cloud is a non-aggregated obstacle, when the first point cloud does not comprise the obstacle aggregation region.

8. A non-transitory computer-readable storage medium, having stored thereon a computer program such that when the computer program is executed by a processor, the computer program causes the processor to perform:

obtaining two frames of laser point clouds acquired by an unmanned driving device as a first point cloud and a second point cloud;

determining an obstacle aggregation region in the first point cloud according to position information of each obstacle in the first point cloud;

determining, according to a position of the obstacle aggregation region, at least one obstacle in the second point cloud that falls within the obstacle aggregation region as at least one aggregated obstacle, and at least one another obstacle, except the at least one aggregated obstacle, in the second point cloud as at least one non-aggregated obstacle; and

for each of the at least one aggregated obstacle,

determining matching parameter values between the aggregated obstacle and each obstacle in the first point cloud by using a group matching rule, and

determining a matching result of the aggregated obstacle according to the matching parameter values between the aggregated obstacle and each obstacle in the first point cloud;

for each of the at least one non-aggregated obstacle,

determining matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using a non-group matching rule, and

determining a matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud, wherein a constraint condition of the group matching rule is stronger than a constraint condition of the non-group matching rule; and

determining an obstacle tracking result according to the matching result of each of the at least one aggregated obstacle and the matching result of each of the at least one non-aggregated obstacle.

9. The non-transitory computer-readable storage medium according to claim 8 , wherein determining the obstacle aggregation region in the first point cloud according to the position information of each obstacle in the first point cloud comprises:

determining a distance between every two obstacles in the first point cloud according to the position information of each obstacle in the first point cloud; and

determining the obstacle aggregation region in the first point cloud according to the distance between every two obstacles in the first point cloud.

10. The non-transitory computer-readable storage medium according to claim 8 , wherein determining the matching parameter values between the aggregated obstacle and each obstacle in the first point cloud by using the group matching rule comprises at least one of:

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a distance between the aggregated obstacle and the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a size of the aggregated obstacle and a size of the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a quantity of points comprised in the aggregated obstacle in the second point cloud and a quantity of points comprised in the obstacle in the first point cloud; or

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a type of the aggregated obstacle and a type of the obstacle in the first point cloud.

11. The non-transitory computer-readable storage medium according to claim 8 , wherein determining the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using the non-group matching rule comprises at least one of:

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a distance between the non-aggregated obstacle and the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a size of the non-aggregated obstacle and a size of the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a quantity of points comprised in the non-aggregated obstacle in the second point cloud and a quantity of points comprised in the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a type of the non-aggregated obstacle and a type of the obstacle in the first point cloud; or

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to an orientation of the non-aggregated obstacle and an orientation of the obstacle in the first point cloud.

12. The non-transitory computer-readable storage medium according to claim 8 , the operations further comprise:

changing the aggregated obstacle to a non-aggregated obstacle in response to the matching result of the aggregated obstacle being not determined.

13. The non-transitory computer-readable storage medium according to claim 8 , wherein determining the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using the non-group matching rule comprises:

determining obstacles in the first point cloud that do not match any aggregated obstacle in the second point cloud as a first obstacle set; and

determining matching parameter values between the non-aggregated obstacle and each of the obstacles in the first obstacle set by using the non-group matching rule; and

determining the matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud comprises:

determining a matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each of the obstacles in the first obstacle set.

14. An unmanned driving device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor is configured to, when executing the program, implement operations comprising:

obtaining two frames of laser point clouds acquired by an unmanned driving device as a first point cloud and a second point cloud;

determining an obstacle aggregation region in the first point cloud according to position information of each obstacle in the first point cloud;

determining, according to a position of the obstacle aggregation region, at least one obstacle in the second point cloud that falls within the obstacle aggregation region as at least one aggregated obstacle, and at least one another obstacle, except the at least one aggregated obstacle, in the second point cloud as at least one non-aggregated obstacle; and

for each of the at least one aggregated obstacle,

determining matching parameter values between the aggregated obstacle and each obstacle in the first point cloud by using a group matching rule, and

determining a matching result of the aggregated obstacle according to the matching parameter values between the aggregated obstacle and each obstacle in the first point cloud;

for each of the at least one non-aggregated obstacle,

determining matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using a non-group matching rule, and

determining a matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud, wherein a constraint condition of the group matching rule is stronger than a constraint condition of the non-group matching rule; and

determining an obstacle tracking result according to the matching result of each of the at least one aggregated obstacle and the matching result of each of the at least one non-aggregated obstacle.

15. The unmanned driving device according to claim 14 , wherein determining the obstacle aggregation region in the first point cloud according to the position information of each obstacle in the first point cloud comprises:

determining a distance between every two obstacles in the first point cloud according to the position information of each obstacle in the first point cloud; and

determining the obstacle aggregation region in the first point cloud according to the distance between every two obstacles in the first point cloud.

16. The unmanned driving device according to claim 14 , wherein determining the matching parameter values between the aggregated obstacle and each obstacle in the first point cloud by using the group matching rule comprises at least one of:

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a distance between the aggregated obstacle and the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a size of the aggregated obstacle and a size of the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a quantity of points comprised in the aggregated obstacle in the second point cloud and a quantity of points comprised in the obstacle in the first point cloud; or

for each obstacle in the first point cloud, determining a matching parameter value between the aggregated obstacle and the obstacle in the first point cloud according to a type of the aggregated obstacle and a type of the obstacle in the first point cloud.

17. The unmanned driving device according to claim 14 , wherein determining the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using the non-group matching rule comprises at least one of:

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a distance between the non-aggregated obstacle and the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a size of the non-aggregated obstacle and a size of the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a quantity of points comprised in the non-aggregated obstacle in the second point cloud and a quantity of points comprised in the obstacle in the first point cloud;

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to a type of the non-aggregated obstacle and a type of the obstacle in the first point cloud; or

for each obstacle in the first point cloud, determining a matching parameter value between the non-aggregated obstacle and the obstacle in the first point cloud according to an orientation of the non-aggregated obstacle and an orientation of the obstacle in the first point cloud.

18. The unmanned driving device according to claim 14 , the operations further comprise:

changing the aggregated obstacle to a non-aggregated obstacle in response to the matching result of the aggregated obstacle being not determined.

19. The unmanned driving device according to claim 14 , wherein determining the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud by using the non-group matching rule comprises:

determining obstacles in the first point cloud that do not match any aggregated obstacle in the second point cloud as a first obstacle set; and

determining matching parameter values between the non-aggregated obstacle and each of the obstacles in the first obstacle set by using the non-group matching rule; and

determining the matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each obstacle in the first point cloud comprises:

determining a matching result of the non-aggregated obstacle according to the matching parameter values between the non-aggregated obstacle and each of the obstacles in the first obstacle set.

20. The unmanned driving device according to claim 14 , the operations further comprise:

determining that each obstacle comprised in the second point cloud is a non-aggregated obstacle, when the first point cloud does not comprise the obstacle aggregation region.

Assignments (2)
CORRECTIVE ASSIGNMENT TO CORRECT THE ASSIGNEE ADDRESS PREVIOUSLY RECORDED ON REEL 058985 FRAME 0822. ASSIGNOR(S) HEREBY CONFIRMS THE ADDRESS IS:ROOM 2106-030, NO.9 WEST NORTH 4TH RING ROAD, HAIDIAN DISTRICT. Recorded Feb 22, 2022
From: XIA, HUAXIA; CAI, SHANBO
To: BEIJING SANKUAI ONLINE TECHNOLOGY CO., LTD.
Reel/Frame 059353/0501 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 11, 2022
From: XIA, HUAXIA; CAI, SHANBO
To: BEIJING SANKUAI ONLINE TECHNOLOGY CO., LTD.
Reel/Frame 058985/0822 →
Priority Claims (1)
CN 202110364576.6 · Apr 6, 2021 · national
Continuity (1)
Related Publication 20220314980A1 · Oct 6, 2022
References Cited (9)
US 20190080183A1 · Li et al. · 2019 [cited by applicant]
CN 104318206A · 2015 [cited by applicant]
CN 109509210A · 2019 [cited by applicant]
CN 110458055A · 2019 [cited by applicant]
CN 112347999A · 2021 [cited by applicant]
CN 112348897A · 2021 [cited by applicant]
CN 112393735A · 2021 [cited by applicant]
State Intellectual Property Office of the People's Republic of China, Office Action and Search Report Issued in Application No. 2021103645766, May 14, 2021, 10 pages. (Submitted with Machine/Partial Translation). [cited by applicant]
Fel Yang et.al“Real-time dynamic obstacle detection and tracking using 3D Lidar”, Sep. 30, 2012, 7 pages. [cited by applicant]
Cited By (1)
US 12,283,105