INTELLIGENT INTERFACE DEVICE FOR GRASPING OF AN OBJECT BY A MANIPULATING ROBOT AND METHOD OF IMPLEMENTING THIS DEVICE

• US 20100172733A1
• Filed: 03/23/2007
• Published: 07/08/2010
• Est. Priority Date: 03/27/2006
• Status: Active Grant

• Current Assignee: Commissariat a L'Energie Atomique (AES Corporation)
• Original Assignee: Commissariat a L'Energie Atomique (AES Corporation)
• Inventors: Leroux, Christophe, Chalubert, Gerard
• Granted Patent: US 8,155,787 B2
• Time in Patent Office: Days
• US Class Current: 414/730
• CPC Class Codes: A61F 4/00 Methods or devices enabling patients or disabled persons to operate an apparatus or a device not forming part of the body operating or control means for prostheses A61F2/48, A61F2/68
• CPC Class Codes: B25J 9/1671 characterised by simulation, either to verify existing program or to create and verify new program, CAD/CAM oriented, graphic oriented programming systems

Abstract

The invention concerns an intelligent interface device for grasping an object comprising:

• • a manipulating robot comprising a hinged arm provided with a clamp at its free end and equipped with at least one camera,
  • a computer with a display screen and an input unit,
  • means of controlling the clamp,
  • means of displaying the video image of the object taken by a camera on the display screen,
  • means of graphically identifying a selection area surrounding the object in this image using the input unit.

The invention also concerns a method for implementing this device.

Claims

1. An intelligent interface device for grasping of an object comprising:

a manipulating robot comprising a hinged arm (20) provided with a clamp (21) at its free end and equipped with at least one camera (22),

a computer (23) with a display screen (25) and an input unit (24),

means of controlling the clamp (21),

means of displaying the video image of the object taken by a camera on the display screen,

characterized in that it also includes;

means of graphically identifying a selection area (35) surrounding the object (19) in this image using the input unit (24), these graphical identification means being piloted by four commands for upward (30), downward (31), leftward (32) and rightward (33) movement to trace this selection area (35) surrounding the object (19), and a validation command (34), and

in that the means of controlling the clamp (21) comprise at least one graphic control button able to be actuated from the display screen and responding to at least one of the following commands;

move the clamp to the left,

move the clamp to the right,

move the clamp downward,

move the clamp upward

move the clamp forward,

move the clamp backward,

open/close the clamp,

turn the clamp clockwise,

turn the clamp counterclockwise,

validate,

cancel,

immediate stop.

2. The device according to claim 1, comprising two cameras forming a stereoscopic video sensor.

3. The device according to claim 1, in which the input unit (24) is a mouse, a head tracking, a contactor, a virtual keyboard, a joystick, or an ocular monitoring or voice synthesis system.

4. The device according to claim 1, in which each command corresponds to a click on an icon shown on the display screen.

5. The device according to claim 1, in which the clamp is equipped with an optical barrier, or a proximity detector.

6. The device according to claim 1, in which the robot is fixed to the arm of a wheelchair.

7. The device according to claim 1, in which the robot is fixed on a mobile platform.

8. A method for implementing an intelligent interface device for grasping of an object comprising:

a manipulating robot comprising a hinged arm (20) provided with a clamp (21) at its free end and equipped with at least one camera (22),

a computer (23) with a display screen (25) and an input unit (24),

characterized in that it comprises the following steps;

bring the object into the field of vision of a camera by controlling the movement of the clamp (step 40), the video image taken by the camera being displayed on the display screen,

identify a selection area around the object using the input unit (step 41),

discriminate between the object and its environment, and estimate the distance between the clamp and the object (step 42),

calculate the baric center of the object in the image (step 43),

calculate a set speed according to the distance to be traveled by the clamp to reach the object (step 44),

move the clamp up to the vicinity of the object (step 45),

move the clamp blindly and close the clamp (step 46),

bring the object back toward the user (step 47).

9. The method according to claim 8, in which the selection area surrounding the object is a rectangular area, a graphical lasso defined by several points chosen by the user, or a closed line traced surrounding the object.

10. The method according to claim 8, in which one uses two cameras forming a stereoscopic video sensor.

11. The method according to claim 10, in which one selects points of interest in the two images coming from the two cameras.

12. The method according to claim 11, in which one pairs these points two by two.

13. The method according to claim 10, in which one corrects the images coming from the two cameras.