Robotic Grasping by Showing: An Experimental Comparison of Two Algorithms

Abstract This paper describes and evaluates the results of two algorithms for robotic grasp formulation on previously unseen 2-D thick laminae. Each algorithm determines the grasp specification from a binary image of the object to be grasped; one is an approach based on Powers’ Perceptual Control Theory and the other is based on hierarchical occupancy arrays. The experimental comparison of the two algorithms involves locating suitable grasps on objects that are difficult to grasp because of their geometry and because of variations in surface roughness and in product size and shape; they are trapezoidal portions of frozen fish coated in breadcrumbs, although for the purposes of this study artificial replica products were used. Grasp sites were located using the two different algorithms, and the quality of the grasps achieved was tested empirically using a ‘pick-up and shake’ routine on an Adept robot to simulate high speed manipulation in a production environment.

[1]  Shahram Payandeh,et al.  On Computing the Friction Forces Associated With Three-Fingered Grasp , 1994, Int. J. Robotics Res..

[2]  Karun B. Shimoga,et al.  Robot Grasp Synthesis Algorithms: A Survey , 1996, Int. J. Robotics Res..

[3]  Mark H. Lee,et al.  Robotic grasping of complex shapes: is full geometrical knowledge of the shape really necessary? , 1995, Robotica.

[4]  Mark H. Lee,et al.  A visually guided robot system for food handling applications , 1994, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.

[5]  Adel L. Ali,et al.  A man-machine approach for robot grasp planning , 1990 .

[6]  William A. Gruver,et al.  Dynamic grasp planning of multifingered robot hands based on asymptotic stability , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[7]  Mark H. Lee,et al.  Applying vision guidance in robotic food handling , 1996, IEEE Robotics Autom. Mag..

[8]  Uri Tasch,et al.  A new representation for a robot grasping quality measure , 1995, Robotica.

[9]  Enrico Pagello,et al.  A heuristic approach to automatic grasp planning for a 3-fingered hand , 1995, J. Intell. Robotic Syst..

[10]  Patrick Olivier,et al.  Matching and tracking using decomposition , 1996 .

[11]  Kevin Warwick,et al.  A survey of multi-fingered robot hands: Issues and grasping achievements , 1993 .

[12]  W. T. Powers.  Behavior, the control of perception , 1973 .

[13]  Giuseppe Casalino,et al.  Contact Force Canonical Decomposition and the Role of Internal Forces in Robust Grasp Planning Problems , 1996, Int. J. Robotics Res..