Improved Stratified Control for Hexapod Robots and Object Manipulation with Finger Relocation

Abstract

The paper deals with the motion design of legged robots and dextrous hands. We show the possibilities and limitation of conventional stratified control approach through the relative simple example of hexapod robot and offer some proposals for a more robust motion planning solution. The precision of the algorithms was improved by step length modification and the applicability was increased by time scaling. The developed software is based on symbolic computation. On the other hand, our fundamental goal is to provide a powerful basic concept for object manipulation with finger relocation in the context of stratified control as an extension of earlier works. The concept focuses on the finger gaiting manipulation (based on finger relocation) to gain some attributes of the nonsmooth object.