Dynamic walking of biped robots with obstacles using predictive controller

Abstract

This paper proposes a control method for walking of biped robots while stepping over large obstacles, which is a big challenge for these robots. To this end, Nonlinear Model Predictive Control (NMPC) is employed. The main advantage of this approach is that there is no need for trajectory planning. In other words, the robot finds the optimum gait length based on stability of the closed-loop system and going over the obstacles. Simulation results show good performance of the proposed method, where the biped can step over a 40 × 15 cm obstacle in the sagittal plane without colliding with it.