Legged robot motion control method, apparatus, and device, and medium

Patent metadata

Field	Value
Publication number	US-12304083-B2
Application number	US-202217969999-A
Country	US
Kind code	B2
Filing date	Oct 20, 2022
Priority date	Sep 7, 2020
Publication date	May 20, 2025
Grant date	May 20, 2025

Abstract

A legged robot motion control method includes: acquiring centroid state data of a spatial path start point and a spatial path end point of a motion path; determining a target landing point of afoot of the legged robot in the motion path based on the spatial path start point and the spatial path end point; determining a change relationship between a centroid position change coefficient and a foot contact force based on the centroid state data; selecting, under constraint of a constraint condition set, a target centroid position change coefficient that meets the change relationship; the constraint condition set including a spatial landing point constraint condition; determining a target motion control parameter according to the target centroid position change coefficient and the target landing point of the foot; and controlling, based on the target motion control parameter, the legged robot to perform motion according to the motion path.

First claim

What is claimed is:

1. A legged robot motion control method, performed by a control device and comprising:
acquiring centroid state data respectively corresponding to a spatial path start point and a spatial path end point of a motion path of a legged robot;
determining a target landing point of a foot of the legged robot in the motion path based on the spatial path start point and the spatial path end point;
determining a change relationship between a centroid position change coefficient and a foot contact force based on the centroid state data respectively corresponding to the spatial path start point and the spatial path end point;
selecting, under constraint of a constraint condition set, a target centroid position change coefficient that meets the change relationship, wherein the constraint condition set comprises at least a spatial landing point constraint condition, and the spatial landing point constraint condition constrains a value of the centroid position change coefficient for the legged robot to reach the target landing point;
determining a target motion control parameter according to the target centroid position change coefficient and the target landing point of the foot; and
controlling, based on the target motion control parameter, the legged robot to perform motion according to the motion path.

2. The method according to claim 1 , wherein the centroid state data corresponding to the spatial path start point comprises a start centroid position of the spatial path start point, and the determining a change relationship between a centroid position change coefficient and a foot contact force based on the centroid state data respectively corresponding to the spatial path start point and the spatial path end point comprises:
obtaining moving duration required by the legged robot to complete the motion path;
determining a plurality of sampling moments from the moving duration, and determining a time interval between each sampling moment and a moment corresponding to the spatial path start point as a time interval corresponding to the sampling moment; and
obtaining a change relationship between the centroid position change coefficient and a foot contact force at each sampling moment according to a time interval corresponding to each sampling moment, the start centroid position, and a relational expression between a centroid position and the foot contact force at each sampling moment.

3. The method according to claim 2 , wherein the obtaining a change relationship between the centroid position change coefficient and a foot contact force at each sampling moment according to a time interval corresponding to each sampling moment, the start centroid position, and a relational expression between a centroid position and the foot contact force at each sampling moment comprises:
determining a centroid angular momentum at each sampling moment according to a body posture of the legged robot at each sampling moment; and
obtaining the change relationship between the centroid position change coefficient and the foot contact force at each sampling moment according to the centroid angular momentum at each sampling moment, the time interval corresponding to each sampling moment, the start centroid position, and the relational expression between the centroid position and the foot contact force at each sampling moment.

4. The method according to claim 2 , wherein the selecting, under constraint of a constraint condition set, a target centroid position change coefficient that meets the change relationship comprises:
determining a target constraint relationship of a centroid position change coefficient at each sampling moment as a target constraint relationship corresponding to each sampling moment according to a foot in contact with a contact surface at each sampling moment, a body posture of the legged robot at each sampling moment, a target landing point of each foot, and the spatial landing point constraint condition; and
selecting the target centroid position change coefficient that meets the target constraint relationship corresponding to each sampling moment and the change relationship between the centroid position change coefficient and the foot contact force at each sampling moment.

5. The method according to claim 4 , wherein the method further comprises:
determining, for each sampling moment according to a rotation range and a length of a joint associated with a foot in contact with the contact surface at the sampling moment, a plurality of positions accessible to the foot in contact with the contact surface; and
fitting the plurality of positions to obtain the spatial landing point constraint condition of the legged robot.

6. The method according to claim 1 , wherein the selecting, under constraint of a constraint condition set, a target centroid position change coefficient that meets the change relationship comprises:
obtaining, under constraint of the constraint condition set, a plurality of groups of candidate results that meet the change relationship; each group of candidate results comprising a centroid position change coefficient; and
minimizing an objective function according to the plurality of groups of candidate results, and determining a target result from the plurality of groups of candidate results; the objective function being a quadratic term constructed according to a correlation quantity of the candidate result.

7. The method according to claim 6 , wherein the objective function comprises at least one of:
a quadratic term related to a change amount of a foot contact force in the motion path;
a quadratic term of a change amount of a centroid position in the motion path;
a quadratic term of a difference between an end centroid position of the spatial path end point in the centroid state data corresponding to the spatial path end point and a centroid position of the spatial path end point in the candidate result;
a quadratic term of a difference between an end centroid speed of the spatial path end point in the centroid state data corresponding to the spatial path end point and a centroid speed of the spatial path end point in the candidate result; or
a quadratic term of a difference between an end centroid acceleration of the spatial path end point in the centroid state data corresponding to the spatial path end point and a centroid acceleration of the spatial path end point in the candidate result.

8. The method according to claim 1 , wherein the constraint condition set further comprises:
a friction constraint condition used for constraining a friction between a foot of a landing point and the contact surface; and the friction between the foot of the landing point and the contact surface is determined according to the foot contact force and a friction coefficient.

9. The method according to claim 1 , wherein the determining a target landing point of the foot in the motion path based on the spatial path start point and the spatial path end point comprises:
collecting a three-dimensional point cloud map of an environment in which the legged robot is currently located; and
determining, from the three-dimensional point cloud map according to a foot of the legged robot in contact with the contact surface in the motion path, a target landing point corresponding to the foot in contact with the contact surface.

10. The method according to claim 1 , wherein the determining a target motion control parameter according to the target centroid position change coefficient and the target landing point of each foot comprises:
determining a centroid position of the legged robot at a moment according to the target centroid position change coefficient; and
determining a joint torque of the le

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Zheng Yu
• Jiang Xinyang
• Chi Wanchao
• Zhang Zhengyou

Classifications

• B62D57/032

  with alternately or sequentially lifted supporting base and legs; with alternately or sequentially lifted feet or skid (B62D57/024 takes precedence) · CPC title

• B25J9/1664 Primary

  characterised by motion, path, trajectory planning · CPC title

• G05D1/0231 Primary

  using optical position detecting means (position-fixing by using electromagnetic waves other than radio waves, e.g. optical position detecting means G01S5/16) · CPC title

• G05D1/0891 Primary

  specially adapted for land vehicles · CPC title