Granted Patent B2

US 12,536,810 · App. 18/460,323 · Granted Jan 27, 2026

Inventors: Zhichao Li (Beijing, CN); Naiyan Wang (Beijing, CN)

Assignee: Beijing OCGen Technology Co., Ltd.

G06V20/58B60W60/001G06T3/40G06T11/00G06V10/776G06T2210/12G06T2210/56

Abstract

Claims (75)

1 . A vehicle control method, comprising:

obtaining a first group of point cloud frames associated with surroundings of a perception system;

generating a hierarchical representation of an object in the surroundings based on the first group of point cloud frames, the hierarchical representation comprising a first representation of the object;

obtaining a second group of point cloud frames associated with the surroundings of the perception system;

determining, from each frame of point cloud in the second group of point cloud frames, a group of points associated with the object and a first subset and a second subset of the group of points;

determining, based on the first subset and the second subset, a first confidence that the object comprises a first part associated with the first subset and a second confidence that the object comprises a second part associated with the second subset;

determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence;

generating a trajectory for the vehicle to travel along at least partly based on the hierarchical representation of the object; and

controlling the vehicle to travel along the generated trajectory,

wherein determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence comprises:

changing at least one of a size and a shape of the first representation in response to one of the first confidence and the second confidence being less than the predetermined threshold and the other of the first confidence and the second confidence being greater than or equal to the predetermined threshold,

wherein generating the trajectory for the vehicle to travel along based on the hierarchical representation of the object comprises:

generating the trajectory for the vehicle to travel along based on the changed hierarchical representation of the object.

2 . The method according to claim 1 , wherein the second group of point cloud frames comprises a plurality of frames of point clouds,

wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset comprises:

for each frame of point cloud in the second group of point cloud frames, determining a first probability that the object comprises the first part based on the first subset,

for each frame of point cloud in the second group of point cloud frames, determining a second probability that the object comprises the second part based on the second subset, and

accumulating a plurality of first probabilities and a plurality of second probabilities, respectively.

3 . The method according to claim 2 , wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset further comprises:

for each frame of point cloud in the second group of point cloud frames, determining a first association probability that the object comprises the first part according to the first subset and the second subset,

for each frame of point cloud in the second group of point cloud frames, determining a second association probability that the object comprises the second part according to the first subset and the second subset, and

accumulating the first association probabilities and the second association probabilities, respectively.

4 . The method according to claim 2 , wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset further comprises:

determining, based on a pre-stored data set, an accuracy rate for the first probability and the second probability, respectively, for each frame of point cloud in the second group of point cloud frames.

5 . The method according to claim 1 , wherein the object is a truck, the first part is a tractor of the truck, and the second part is a trailer of the truck.

6 . An electronic device, comprising: a processor and a memory, wherein the memory stores program instructions executable by the processor to implement a vehicle control method, comprising:

obtaining a first group of point cloud frames associated with surroundings of a perception system;

generating a hierarchical representation of an object in the surroundings based on the first group of point cloud frames, the hierarchical representation comprising a first representation of the object;

obtaining a second group of point cloud frames associated with the surroundings of the perception system;

determining, from each frame of point cloud in the second group of point cloud frames, a group of points associated with the object and a first subset and a second subset of the group of points;

determining, based on the first subset and the second subset, a first confidence that the object comprises a first part associated with the first subset and a second confidence that the object comprises a second part associated with the second subset;

determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence;

generating a trajectory for the vehicle to travel along at least partly based on the hierarchical representation of the object; and

controlling the vehicle to travel along the generated trajectory,

wherein determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence comprises:

changing at least one of a size and a shape of the first representation in response to one of the first confidence and the second confidence being less than the predetermined threshold and another being greater than or equal to the predetermined threshold,

wherein generating the trajectory for the vehicle to travel along based on the hierarchical representation of the object comprises:

generating the trajectory for the vehicle to travel along based on the changed hierarchical representation of the object.

7 . The electronic device according to claim 6 , wherein the second group of point cloud frames comprises a plurality of frames of point clouds,

wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset comprises:

for each frame of point cloud in the second group of point cloud frames, determining a first probability that the object comprises the first part based on the first subset,

for each frame of point cloud in the second group of point cloud frames, determining a second probability that the object comprises the second part based on the second subset, and

accumulating a plurality of first probabilities and a plurality of second probabilities, respectively.

8 . The electronic device according to claim 7 , wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset further comprises:

for each frame of point cloud in the second group of point cloud frames, determining a first association probability that the object comprises the first part according to the first subset and the second subset,

for each frame of point cloud in the second group of point cloud frames, determining a second association probability that the object comprises the second part according to the first subset and the second subset, and

accumulating the first association probabilities and the second association probabilities, respectively.

9 . The electronic device according to claim 7 , wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset further comprises:

determining, based on a pre-stored data set, an accuracy rate for the first probability and the second probability, respectively, for each frame of point cloud in the second group of point cloud frames.

10 . The electronic device according to claim 6 , wherein the object is a truck, the first part is a tractor of the truck, and the second part is a trailer of the truck.

11 . A non-transitory computer-readable storage medium, wherein the computer-readable storage medium stores program instructions executable by a processor to implement a vehicle control method, comprising:

obtaining a first group of point cloud frames associated with surroundings of a perception system;

generating a hierarchical representation of an object in the surroundings based on the first group of point cloud frames, the hierarchical representation comprising a first representation of the object;

obtaining a second group of point cloud frames associated with the surroundings of the perception system;

determining, from each frame of point cloud in the second group of point cloud frames, a group of points associated with the object and a first subset and a second subset of the group of points;

determining, based on the first subset and the second subset, a first confidence that the object comprises a first part associated with the first subset and a second confidence that the object comprises a second part associated with the second subset;

determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence;

generating a trajectory for the vehicle to travel along at least partly based on the hierarchical representation of the object; and

controlling the vehicle to travel along the generated trajectory,

wherein determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence comprises:

changing at least one of a size and a shape of the first representation in response to one of the first confidence and the second confidence being less than the predetermined threshold and another being greater than or equal to the predetermined threshold,

wherein generating the trajectory for the vehicle to travel along based on the hierarchical representation of the object comprises:

generating the trajectory for the vehicle to travel along based on the changed hierarchical representation of the object.

12 . The non-transitory computer-readable storage medium according to claim 11 , wherein the second group of point cloud frames comprises a plurality of frames of point clouds,

wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset comprises:

for each frame of point cloud in the second group of point cloud frames, determining a first probability that the object comprises the first part based on the first subset,

for each frame of point cloud in the second group of point cloud frames, determining a second probability that the object comprises the second part based on the second subset, and

accumulating a plurality of first probabilities and a plurality of second probabilities, respectively.

13 . The non-transitory computer-readable storage medium according to claim 12 , wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset further comprises:

for each frame of point cloud in the second group of point cloud frames, determining a first association probability that the object comprises the first part according to the first subset and the second subset,

for each frame of point cloud in the second group of point cloud frames, determining a second association probability that the object comprises the second part according to the first subset and the second subset, and

accumulating the first association probabilities and the second association probabilities, respectively.

14 . The non-transitory computer-readable storage medium according to claim 12 , wherein determining, based on the first subset and the second subset, the first confidence that the object comprises the first part associated with the first subset and the second confidence that the object comprises the second part associated with the second subset further comprises:

determining, based on a pre-stored data set, an accuracy rate for the first probability and the second probability, respectively, for each frame of point cloud in the second group of point cloud frames.

15 . The non-transitory computer-readable storage medium according to claim 11 , wherein the object is a truck, the first part is a tractor of the truck, and the second part is a trailer of the truck.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (9)