Gait Planning for Underactuated Bipedal Robot on Discontinuous Ground Based on Particle Swarm Optimization

This study develops a gait planning method for underactuated bipedal robot on discontinuous ground based on Particle Swarm Optimization (PSO). First, we introduce an underactuated bipedal walking gait planning method based on Center-of-Mass (CoM) motion state. Second, based on human walking characteristics, we propose a gait planning method to realize underactuated bipedal walking on discontinuous ground. We change the step length and the ideal tracking speed of CoM, in order to eliminate the influence of the ground discontinuity on walking stability. By using PSO, we obtain the ideal tracking speed of CoM for different gait. Finally, the effectiveness of the proposed method is validated with a numerical simulation and a physical experiment. The experimental results show that underactuated bipedal walking can be realized on discontinuous ground by using the proposed method to control robot step length and the ideal tracking speed of CoM.