Cyclic gait planning and control of a five-link biped robot with four actuators during single support and double support phases

Abstract

This paper focuses on cyclic gait planning and motion control of a five-link biped robot with four actuators. The biped walking consists of two alternative phases of motion: single support phase (SSP) and double support phase (DSP). The biped robot is under-actuated and over-actuated during the SSP and DSP, respectively. Because of the underactuation problem in the SSP, time trajectories cannot be obtained directly and control motion is difficult. Hence, a new algorithm is proposed for cyclic gait planning during the successive single and double support phases. Two control laws are proposed for the control motion during the SSP and the DSP. In the SSP, the control law is developed using a predictive control algorithm and a time optimal control is proposed for the DSP. Attraction region and convergence to the cyclic gait are studied. The results demonstrate the effectiveness of the proposed method in the biped motion.