Processing Sensor Images for Grasping Disassembly Objects with a Parallel-Jaw Gripper

Abstract

In this paper, the problem of planning suitable grasps for objects of unknown shape especially in the context of disassembling electronic devices is addressed. An autonomous robot system equipped with a parallel-jaw gripper and a laser range finder is employed for acquisition of adequate object information as well as for disassembly operations. The respective objects to be disassembled are enclosed in the casing of the electronic device, and often they are hardly accessible for grasping. In this work, a method for searching and localizing suitable gripping points on arbitrarily shaped objects is presented using sensor images acquired by a laser range finder. For sensor image processing the range data are represented in a 2D gray level image without losing 3D information. Experimental results prove the efficiency of the system for a simple "block-in-a-box" task as well as for a more complex problem of grasping the objective module inside a camcorder. 1 Introduction Considering the nega...