Proposal of Augmented Linear Inverted Pendulum model for bipedal gait planning

Abstract

In this paper, we propose a new model called Augmented Linear Inverted Pendulum (ALIP) in which an augmented function F is added to the dynamic equation of the linear inverted pendulum. The purpose of adding the function F is to modify/adjust the inverted pendulum dynamics in such a way that disturbance caused by un-modeled dynamics (legs, arms, etc.) can be compensated or minimized. By changing the key parameters of the augmented function we can easily modify the inverted pendulum dynamics. The desired walking motion with maximized stability margin is achieved by optimizing the key parameters using genetic algorithm. The disturbance created by the un-modeled dynamics is minimized because full robot dynamics is considered in the optimization process. Simulations results show that the walking gait obtained using the proposed method is more stable than that obtained using the Linear Inverted Pendulum Mode (LIPM).