Reconstruction, retargetting, tracking, and estimation of pose of articulated systems

• US 8,467,904 B2
• Filed: 12/21/2006
• Issued: 06/18/2013
• Est. Priority Date: 12/22/2005
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Dariush, Behzad
• Primary Examiner(s): BEHNCKE, CHRISTINE M
• Application Number:
  US11/614,930
• Publication Number:
  US 20070162164A1
• Time in Patent Office: 2,371 Days
• Field of Search: 700/245, 700/246, 700/257, 700/259, 700/262, 382/153, 382/103, 382/115, 382/107, 703/11, 345/474, 345/473, 702/150, 901/3, 901/50
• US Class Current: 700/262
• CPC Class Codes: B25J 9/1602 characterised by the control system, structure, architecture
• CPC Class Codes: G05B 2219/39062 Calculate, jacobian matrix estimator
• CPC Class Codes: G05B 2219/39219 Trajectory tracking
• CPC Class Codes: G05B 2219/39381 Map task, application to behaviour, force tracking, singularity to motion to actuator
• CPC Class Codes: G05B 2219/40264 Human like, type robot arm
• CPC Class Codes: G05B 2219/40476 Collision, planning for collision free path
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80

Abstract

A control system and method generate joint variables for motion or posing 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 or posing as replicated by the target system.

Claims

1. A computer-implemented method for controlling the pose of a physical articulated system, the method comprising:

observing a plurality of task descriptors for a position and an orientation of members of a source system comprising human features observed from images of human motion;

generating a position error between a computed position and the position of the members of the source system;

generating an orientation error between a computed orientation and the orientation of the members of the source system;

generating, using a non-transitory computer readable storage medium, a control command according to a control law based on the position error, orientation error, and kinematic constraints;

constructing, using the non-transitory computer readable storage medium, a weighted matrix based on the generated control command and the kinematic constraints of members of the physical articulated system;

generating, using the non-transitory computer readable storage medium, joint variables for positioning and orienting the members of the physical articulated system while enforcing the kinematic constraints of the members by applying the weighted matrix to the observed task descriptors, the number of task descriptors being less than a degrees of freedom of movement of the source system; and

storing the generated joint variables in the non-transitory computer readable storage medium.

2. The method of claim 1, further comprising:

generating constraints on positions of the members of the physical articulated system, wherein the weighted matrix is constructed based at least in part on the constraints.

3. The method of claim 2 wherein the observing comprises second order closed loop inverse kinematics.

4. The method of claim 3 wherein the observing further comprises regularizing an inverse Jacobian matrix of a motion of the source system.

5. The method of claim 1 further comprising:

generating a balance control command in response to the generated joint variables.

6. The method of claim 1 wherein the observing comprises first order closed loop inverse kinematics.

7. The method of claim 6 wherein the observing further comprises regularizing an inverse Jacobian matrix of the position of the members of the source system.

8. The method of claim 1 further comprising:

calculating forward kinematics of the joint variables to generate said computed position and computed orientation.

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

10. The method of claim 1 wherein the kinematic constraints comprise singularities arising from mathematics of the generating.

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

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

13. A system for positioning a physical articulated system, the system comprising:

a first control system for observing a plurality of task descriptors for a position and an orientation of members of a source system, comprising human features observed from images of human motion;

a second control system for

generating a position error between a computed position and the position of the members of the source system;

generating an orientation error between a computed orientation and the orientation of the members of the source system;

generating a control command according to a control law based on the position error, orientation error, and kinematic constraints;

constructing a weighted matrix based on the generated control command and the kinematic constraints of members of the physical articulated system, and

generating joint variables for positioning and orienting the members of the physical articulated system while enforcing the kinematic constraints of the members by applying the weighted matrix to the observed task descriptors, the number of task descriptors being less than a degrees of freedom of movement of the source system; and

a non-transitory computer readable storage medium for storing the generated joint variables.

14. The system of claim 13 wherein the second control system is further configured for generating constraints on positions of the members of the physical articulated system, wherein the weighted matrix is constructed based at least in part on the constraints.

15. The system of claim 13 wherein the second control system is further configured for generating a balance control command in response to the generated joint variables.

16. The system of claim 13 wherein the observing comprises first order closed loop inverse kinematics.