Two motion planning approaches for six-legged robot

Abstract

This paper concerns motion planning on a kinematic model of a six-legged robot from two distinct aspects. The system is subjected to intermittent nonholonomic constraints making the robot model stratified, which requires additional considerations in motion planning. An improved version of stratified control is compared with an approach exploiting differential flatness of a restricted subsystem. The main motivation in both cases is to lead the robot along a reference trajectory that assists in resolving the obstacle avoidance problem.