Systems and methods for planning a robot grasp based upon a demonstrated grasp

• US 9,321,176 B1
• Filed: 04/01/2015
• Issued: 04/26/2016
• Est. Priority Date: 04/01/2014
• Status: Active Grant

• Current Assignee: University of South Florida (State University System of Florida)
• Original Assignee: University of South Florida (State University System of Florida)
• Inventors: Sun, Yu, Lin, Yun
• Primary Examiner(s): Anwari, Maceeh
• Application Number:
  US14/676,050
• Time in Patent Office: 391 Days
• Field of Search: 700/245
• US Class Current: 1/1
• CPC Class Codes: B25J 15/0009 comprising multi-articulated fingers, e.g. resembling a human hand
• CPC Class Codes: B25J 15/0028 with movable, e.g. pivoting gripping jaw surfaces
• CPC Class Codes: B25J 15/103 for gripping the object in three contact points
• CPC Class Codes: B25J 9/1612 characterised by the hand, wrist, grip control
• CPC Class Codes: B25J 9/1656 characterised by programming, planning systems for manipulators
• CPC Class Codes: G05B 2219/36422 During teaching shut off, disable motor to move arm easy
• CPC Class Codes: G05B 2219/39526 Three fingers used, thumb, index, middle finger for lateral precision
• CPC Class Codes: G05B 2219/39546 Map human grasps to manipulator grasps
• CPC Class Codes: Y10S 901/39 Jaw structure

Abstract

In one embodiment, planning a robot grasp of an object includes determining a grasp type that would be used by a human being to grasp the object, determining a position and orientation of the human being'"'"'s thumb relative to the object, and planning the robot grasp based upon the determined grasp type and thumb position and orientation.

Claims

1. A method for planning a robot grasp of an object using a computer system, the method comprising:

the computer system determining a grasp type that would be used by a human being to grasp the object;

the computer system determining a position and orientation of the human being'"'"'s thumb relative to the object; and

the computer system planning the robot grasp based upon the determined grasp type and thumb position and orientation by reducing a size of a grasp search space based upon the determined grasp type and thumb position and orientation without considering the positions and orientations of the human being'"'"'s other fingers, and searching the search space for a robot grasp that maximizes a quality function.

2. The method of claim 1, wherein determining a grasp type comprises determining the grasp type using a tracking system that tracks the human being'"'"'s hand when the human being demonstrates grasping of the object.

3. The method of claim 2, wherein the tracking system tracks the human being'"'"'s hand with a data glove.

4. The method of claim 1, wherein determining a position and orientation of the human being'"'"'s thumb comprises determining the position and orientation using a tracking system that tracks the human being'"'"'s thumb when the human being demonstrates grasping of the object.

5. The method of claim 4, wherein the tracking system comprises an optical tracking system that optically tracks the human being'"'"'s thumb.

6. The method of claim 1, wherein determining a grasp type comprises receiving a selection input by the human being that identifies a grasp type the human being would use to grasp the object.

7. The method of claim 6, wherein determining a position and orientation of the human being'"'"'s thumb comprises receiving a selection input by the human being that identifies the position and orientation of the thumb relative to a model of the object that is displayed to the human being.

8. The method of claim 1, wherein planning the robot grasp further comprises correlating the determined grasp type to a robot grasp that the robot can perform.

9. The method of claim 8, wherein planning the robot grasp further comprises mapping the position and orientation of the thumb to a hand of the robot.

10. The method of claim 9, further comprising not mapping a position and orientation of any of the human being'"'"'s other fingers to the robot hand to simplify the mapping process.

11. A system for planning a robot grasp of an object, the system comprising:

a non-transitory computer-readable memory that stores;

logic configured to determine a grasp type a human being would use to grasp the object;

logic configured to determine a position and orientation of the human being'"'"'s thumb relative to the object; and

logic configured to plan the robot grasp based upon the determined grasp type and thumb position and orientation by reducing a size of a grasp search space based upon the determined grasp type and thumb position and orientation without considering the positions and orientations of the human being'"'"'s other fingers, and searching the search space for a robot grasp that maximizes a quality function.

12. The system of claim 11, wherein the logic configured to determine a grasp type comprises logic configured to determine the grasp type from data collected with a tracking system that tracks the human being'"'"'s hand when a demonstrated grasp is performed.

13. The system of claim 11, wherein the logic configured to determine a position and orientation of the human being'"'"'s thumb comprises logic configured to determine a position and orientation from data collected with a tracking system that tracks the human being'"'"'s thumb when a demonstrated grasp is performed.

14. The system of claim 11, wherein the logic configured to determine a grasp type comprises logic configured to receive a selection input by the human being that identifies a grasp type the human being would use to grasp the object.

15. The system of claim 14, wherein the logic configured to determine a position and orientation of the human being'"'"'s thumb comprises logic configured to receive a selection input by the human being that identifies a position and orientation of the thumb relative to a model of the object that is displayed to the human being.

16. The system of claim 11, wherein the logic configured to plan the robot grasp is configured to correlate the determined type of grasp with a type of robot grasp that the robot can perform.

17. The system of claim 16, wherein the logic configured to plan the robot grasp is configured to map the determined position and orientation of the thumb to the robot but not map a position and orientation of any of the human being'"'"'s other fingers to the robot hand to simplify the mapping process.

18. A method for planning a robot grasp of an object performed by a computer system, the method comprising:

the computer system determining a position and orientation of a human being'"'"'s thumb relative to the object; and

the computer system planning the robot grasp based upon the determined thumb position and orientation, wherein the planning comprises reducing a size of a grasp search space based upon the determined thumb position and orientation, searching the search space for a robot grasp that maximizes a quality function, and mapping the position and orientation of the thumb to a hand of a robot without mapping a position and orientation of any of the human being'"'"'s other fingers to the robot hand to simplify the mapping process.