Control Of Robots From Human Motion Descriptors

• US 20070255454A1
• Filed: 04/12/2007
• Published: 11/01/2007
• Est. Priority Date: 04/27/2006
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Dariush, Behzad
• Granted Patent: US 8,924,021 B2
• Time in Patent Office: Days
• US Class Current: 700/245
• CPC Class Codes: G05B 2219/40391 Human to robot skill transfer
• CPC Class Codes: G05B 2219/40495 Inverse kinematics model controls trajectory planning and servo system
• CPC Class Codes: G06N 3/008 based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour
• CPC Class Codes: Y10S 901/03 Teaching system
• CPC Class Codes: Y10S 901/46 Sensing device

Abstract

A control system and method generate torque comments for motion of a target system in response to observations of a source system. Constraints and balance control may be provided for more accurate representation of the motion as replicated by the target system.

Claims

1. A method for controlling movement of members of an articulated system, the method comprising:

tracking motion of a source system;

enforcing constraints on movement of the members of the articulated system;

receiving sensed motion information of the members of the articulated system; and

generating torque commands for controlling movement of the members of the articulated system in response to the tracking motion, the sensed motion information, and the constraints.

2. The method of claim 1 further comprising:

generating a balance control command in response to the sensed motion information.

3. The method of claim 1 wherein the tracking includes inverse kinematics.

4. The method of claim 3 wherein the tracking further comprises regularizing an inverse Jacobian matrix used to describe the differential kinematics of variables in the articulated system.

5. The method of claim 1 wherein the generated torque commands includes:

generating an error between a function of the torque command and the sensed motion information; and

generating the torque commands according to a control law in response to the error, the sensed motion information, and the constraints.

6. The method of claim 1 further comprising:

calculating forward kinematics of the sensed motion information to generate said function of the command.

7. The method of claim 1 wherein generating the torque commands further comprises:

calculating position and motion commands in response to tracking motion;

calculating torque in response to the position and motion commands and sensed motion information.

8. The method of claim 7 wherein generating the torque commands further comprises calculating torque in response to the position and motion commands and sensed motion information and motion characteristics of said articulated system.

9. The method of claim 8 wherein calculating torque command further comprises decoupling dynamically joint space of the motion characteristics of said articulated system.

10. The method of claim 8 wherein calculating torque command further comprises controlling joints of the articulated system independently of each other.

11. The method of claim 1 wherein generating the torque commands further comprises:

calculating position and motion commands in response to tracking motion and the sensed motion information;

calculating torque in response to the position and motion commands and sensed motion information and motion characteristics of said articulated system.

12. The method of claim 1 wherein the constraints include singularities arising from limitations of performing tasks due to insufficient degrees of freedom for executing the tasks.

13. The method of claim 1 wherein the constraints include singularities arising from mathematics of the generating a control command.

14. The method of claim 1 wherein the articulated system is a human model.

15. The method of claim 1 wherein the articulated system is a robot.

16. The method of claim 1 wherein the articulated system is a exoskeleton apparatus.

17. The method of claim 1 wherein enforcing constraints includes:

modulating time between sample times of generating torque commands if the movement of a member of the articulated system has a velocity above a threshold.

18. The met hod of claim 1 wherein enforcing constraints includes:

prioritizing tasks.

19. The method of claim 1 wherein the source motion comprises human features obtained from images or data.

20. A system for controlling movement of members of an articulated system, the system comprising:

a tracking system for tracking motion of a source system;

a constraint system for enforcing constraints on movement of the members of the articulated system;

an input for receiving sensed motion information of the members of the articulated system and

a control system for generating torque commands for controlling movement of the members of the articulated system in response to the tracking motion, the sensed Motion information, and the constraints.

21. A method of tracking motion of a human for replication by an articulated system, the method comprising:

assigning human motion primitives to portions of the human;

detecting motion of the portions of the human represented by the human motion primitives;

generating torque commands for the articulated system to replicate motion of the human in response to the detected motion;

adjusting the torque commands for the replication motion in response to physical limitations of the articulated system for performing motions.

22. The method of claim 21 wherein the human motion primitives have a different dimension than the degrees of freedom of movement of the human.

23. The method of claim 21 wherein the physical limitations include joint limits, velocity limits, and collision avoidance.

24. The method of claim 21 further comprising adjusting the torque commands for the replication motion based on balancing the articulated system during the motion.

25. The method of claim 21 wherein the physical limitations include singularities.

26. The method of claim 21 further comprising adjusting the torque commands in response to position error and orientation error of the assigned task descriptors and sensed position information of the articulated system.

27. The method of claim 21 wherein the articulated system is a human model.

28. The method of claim 21 wherein the articulated system is a robot.

29. The method of claim 21 further comprising:

determining collision points of the articulated system; and

adjusting the torque commands for the replicated motion in response to the collision points.

30. The method of claim 29 wherein the collision points are passive and reactive collision points.

31. A system of tracking motion of a human for replication by an articulated system, the system comprising:

a first control system for assigning human motion primitives to portions of the human;

a second control system for detecting motion of the portions of the human represented by the human motion primitives;

a third control system for generating torque commands for the articulated system to replicate motion of the human in response to the detected motion;

a fourth control system for adjusting the torque commands for the replication motion in response to physical limitations of the articulated system for performing motions.

32. A method of tracking motion of a human for replication by an articulated system, the method comprising:

assigning human motion primitives to portions of the human;

detecting motion of the portions of the human represented by the human motion primitives;

generating torque commands for the articulated system to replicate motion of the human in response to the detected motion;

adjusting the torque commands for the replication motion in response to a preference of human motion primitives of the articulated system for performing motions.

33. The method of claim 32 wherein the adjusting the commands includes assigning preferences to the human motion primitives based on a priority of the task.

34. The method of claim 32 wherein the adjusting the commands includes assigning preferences to the human motion primitives based on a weighting of the human motion primitives.

35. A system of tracking motion of a human for replication by an articulated system, the system comprising:

a first control system for human motion primitives;

a second control system for detecting motion of the portions of the human represented by the human motion primitives;

a third control system generating torque commands for the articulated system to replicate motion of the human in response to the detected motion; and

a fourth control system for adjusting the torque commands for the replication motion in response to a preference of human motion primitives of the articulated system for performing motions.