Real Time Computer Control of a Quadruped Walking Robot

Abstract

This paper presents a description of the hardware and software systems used to achieve real-time computer control of locomotion in an experimental quadruped walking robot. This robot possesses three independently powered joints in each of its four legs. Each joint provides position and rate feedback to a microprocessor-based controller. To control the movement of the leg so that the leg is placed on the desired foothold position, nonlinear feedback is first applied for decoupling, and then Model-Reference-Adaptive-Control is used to control the motion and position of the leg. Experimental results show that the proposed control algorithm can effectively control the leg.