Online Walking Pattern Generation and System Software of Humanoid BHR-2

Abstract

This paper proposed a method of online walking pattern generation based on key parameters of off-line typical walking patterns for humanoid robot. The key parameters include hip parameters, step length, walking cycle and so on. The offline walking pattern generation method is based on ZMP. The design of the system software of humanoid BHR-2 is also described in this paper, in order to ensure the real task can be finished on time, the process of online walking patterns generation is placed outside the processor which used for real-time control. The effectiveness of the proposed method is confirmed by simulations and experiments with our developed humanoid robot with 32 DOF