A Trajectory Generation Method for Humanoid Robots Walking: Optimal Force Pattern Method

Abstract

From the viewpoint of support force, a new gait planning method for Humanoid robots is discussed. A 3D force-pattern based inverted pendulum model (3D-FPBIPM) is built to generate the trajectory for humanoid walking. In the model, support force of leg of humanoid robots is defined by a parametric force pattern function. A nonlinear optimizer is used to search the parameters online for each step. With a series of optimal support force patterns, a stable COM trajectory at a given walking speed is generated by a Runge-Kutta solver. Stride period of the planned gait is very close to the test data and other researcher’s result. The trajectory generated by this method is also compared with the one which is planned by the traditional linear inverted pendulum model. The results show that the proposed method can generate more natural and energy saving walking gait. Finally, the effectiveness of the proposed method is illustrated by simulation examples.