Fast C-shape grasping for unknown objects

Abstract

Grasping of unknown objects with neither appearance data nor object models given in advance is very important for robots that work in an unfamiliar environment. In this paper, we propose an original fast grasping algorithm for unknown objects. The geometry of the under-actuated gripper is approximated as a C-shape, which is used to fit the point cloud of the target object to find a suitable grasp. In order to make the robot arm quickly execute the grasp found by the grasping algorithm, we made a comparison of the popular online motion planners. The motion planner with the highest solved runs, lowest computing time and the shortest path length is chosen to execute the grasp action. Simulations and experiments on a UR5 robot arm and an under-actuated gripper are used to examine the performance of the grasping algorithm, and successful results are obtained.