Method and apparatus for control robot to perform somersault motion, and robot

Patent metadata

Field	Value
Publication number	US-12533811-B2
Application number	US-202318309263-A
Country	US
Kind code	B2
Filing date	Apr 28, 2023
Priority date	May 31, 2021
Publication date	Jan 27, 2026
Grant date	Jan 27, 2026

Abstract

A method for controlling a robot to perform a somersault motion enables the robot, when facing an obstacle such as a hole, ditch, or ravine, to cross the obstacle by performing a somersault motion. The robot includes a wheel-leg portion and a base portion connected via a joint. The robot receives a motion instruction and controls a torque of the joint to allow the robot to perform the somersault. During a takeoff phase of the somersault, a center of mass of the wheel-leg portion is lower than a center of mass of the base portion. During a flight phase, the center of mass of the wheel-leg portion is higher than the center of mass of the base portion at some point in time. During the landing phase, the center of mass of the wheel-leg portion is lower than the base portion. This can enable the robot to avoid obstacles.

First claim

What is claimed is:

1 . A method for controlling a robot, the method comprising:
receiving a motion instruction for the robot, wherein the robot comprises a wheel-leg portion and a base portion connected to the wheel-leg portion via a joint; and controlling a torque of the joint according to the motion instruction to allow the robot to perform a somersault motion through a takeoff phase, a flight phase, and a landing phase; wherein:
during the takeoff phase of the somersault motion, a center of mass of the wheel-leg portion is lower than a center of mass of the base portion,
during the flight phase of the somersault motion, the center of mass of the wheel-leg portion is higher than the center of mass of the base portion at some point in time, and
during the landing phase of the somersault motion, the center of mass of the wheel-leg portion is lower than the center of mass of the base portion;
wherein the controlling the torque of the joint, during the takeoff phase, comprises:
controlling the joint and an active wheel in the wheel-leg portion to allow a central axis of the robot to be inclined to a horizontal direction corresponding to a takeoff direction;
controlling the active wheel to allow the robot to accelerate in a horizontal direction corresponding to the takeoff direction; and
controlling the joint to allow the robot to accelerate in a vertical direction and in the horizontal direction corresponding to the takeoff direction; and
wherein the controlling the torque of the joint, during the flight phase, comprises:
controlling the joint to allow the robot to rotate in a rotational direction corresponding to the takeoff direction, wherein the center of mass of the wheel-leg portion is higher than the center of mass of the base portion at some point in the flight phase.

2 . The method according to claim 1 , wherein the controlling the torque of the joint, during the flight phase, further comprises:
controlling the joint to shorten a distance between the center of mass of the wheel-leg portion and the center of mass of the base portion; and controlling the joint to increase the distance between the center of mass of the wheel-leg portion and the center of mass of the base portion.

3 . The method according to claim 1 , wherein the robot further comprises an additional component, the additional component and the base portion being connected at an additional joint, and wherein controlling the torque of the joint further comprises:
controlling the additional joint to allow the additional component to rotate in the rotational direction corresponding to the takeoff direction.

4 . The method according to claim 3 , wherein the additional component comprises a link connected to an additional active wheel, and wherein during a rotation of the additional component, an angle between a central axis of the link and a perpendicular bisector of the base portion is gradually increased then gradually decreased along the rotational direction.

5 . The method according to claim 1 , wherein controlling the torque of the joint further comprises:
during the landing phase of the somersault motion, controlling the joint and an active wheel in the wheel-leg portion so the robot is in a balanced and upright state.

6 . The method according to claim 1 , wherein controlling the torque of the joint further comprises:
controlling the joint to allow a connecting line between a center of an active wheel in the wheel-leg portion and a center of the base portion to intersect with a perpendicular bisector of the base portion.

7 . The method according to claim 1 , wherein the motion instruction comprises target motion parameters corresponding to different times in the somersault motion of the robot.

8 . The method according to claim 7 , wherein the motion instruction is obtained according to whole-body dynamics planning, the whole-body dynamics planning corresponding to a dynamic model and a set of constraint equations of the robot, and wherein the dynamic model and the set of constraint equations of the robot are associated with a somersault initial target state and a somersault final target state.

9 . The method according to claim 7 , wherein controlling the torque of the joint further comprises:
determining a control signal for controlling the joint based on whole-body dynamics control and according to parameter differences between actual motion parameters of the robot in the somersault motion and the target motion parameters, to control the actual motion parameters of the robot towards the target motion parameters.

10 . The method according to claim 9 , wherein the whole-body dynamics control is based on a dynamic model corresponding to the robot, and the dynamic model is equivalent to one of the following:
a variable-leg-length wheeled inverted pendulum model, the variable-leg-length wheeled inverted pendulum model comprising: a virtual leg with a variable leg length, a virtual active wheel connected to a first end of the virtual leg, and a virtual base portion connected to a second end of the virtual leg, the virtual leg being equivalent to the joints and links in the wheel-leg portion, the virtual active wheel being equivalent to an active wheel in the wheel-leg portion, and the virtual base portion being equivalent to the base portion; or a plane-with-floating-base whole-body dynamic model, the plane-with-floating-base whole-body dynamic model comprising: a virtual front leg, a virtual back leg, a virtual active wheel connected to a first end of the virtual front leg and a first end of the virtual back leg, and a virtual base portion connected to a second end of the virtual front leg and a second end of the virtual back leg, the virtual front leg and the virtual back leg being equivalent to the joints and links in the wheel-leg portion, the virtual active wheel being equivalent to an active wheel in the wheel-leg portion, and the virtual base portion being equivalent to the base portion.

11 . The method according to claim 10 , wherein the variable-leg-length wheeled inverted pendulum model and the plane-with-floating-base whole-body dynamic model further comprise a virtual tail, wherein the virtual tail is equivalent to an additional component.

12 . The method according to claim 10 , wherein,
the variable-leg-length wheeled inverted pendulum model is constructed based on at least one of the following: a center position of the virtual base portion, a rotated angle of the virtual active wheel, a leg length of the virtual leg, a pitch angle of the virtual base portion, an execution torque of the virtual active wheel, a force for changing the leg length of the virtual leg, a friction force, or a support force of a bearing surface to the variable-leg-length wheeled inverted pendulum model; and the plane-with-floating-base whole-body dynamic model is constructed based on at least one of the following: a center position of the virtual base portion, a rotated angle of the virtual active wheel, a pose of the virtual front leg, a pose of the virtual back leg, a pitch angle of the virtual base portion, an execution torque of the virtual active wheel, a rotational torque at a connection between the virtual leg and the virtual base portion, a friction force, or a support force of the bearing surface to the plane-with-floating-base whole-body dynamic model.

13 . The method according to claim 10 , wherein the whole-body dynamics control is further based on a set of constraint equations corresponding to the robot, and the set of constraint equations comprises at least one of the following:
a motion state equation, configured to constrain a motion bet

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Jiang Xinyang
• Zheng Yu
• Wang Shuai
• Wang Haitao
• Xu Jiafeng

Classifications

• B62D57/02 Primary

  with ground-engaging propulsion means, e.g. walking members · CPC title

• B25J9/1633

  compliant, force, torque control, e.g. combined with position control · CPC title

• B25J9/161

  Hardware, e.g. neural networks, fuzzy logic, interfaces, processor · CPC title

• B25J9/1605

  Simulation of manipulator lay-out, design, modelling of manipulator · CPC title

• B25J9/1679 Primary

  characterised by the tasks executed · CPC title