Performance Evaluation and Design Criteria for Manipulation Planning based on Randomized Approach
Masahito Yashima, Takahiro HONDA · TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series C · 2005
This paper presents a planner for manipulating an object by a multi-fingered hand. Inspired by random motion planning for collision-free problems, we propose a novel algorithm for manipulation planning based on randomized approach in order to explore the state space of a manipulation system, and generate subgoals rapidly and uniformly. In addition, we discuss performance evaluation of the proposed planner by executing several simulations. Based on the results, we show design criteria for the randomized planning in order to get solutions efficiently.