An energy efficient gait trajectory planning algorithm for a seven linked biped robot using movement elements

Abstract

In this paper, a computational algorithm is introduced to generate energy efficient gait patterns. Each space cycle of gait has been decomposed to several phases, and boundaries of each phase are called via points. The algorithm uses six universal movement elements [1] together with kinematic information of the physical system at via points. We also developed an optimization procedure to obtain energy efficient kinematic information for the via points and called the new algorithm "optimized-COMAP". The algorithms was used to generate gait patterns for a human like, seven link biped robot. For the joint trajectories of the lower extremities of a optmized-COMAP have been used. Then a desired ZMP trajectory restricted into two upper and lower boundaries has been used to generate upper-body trajectory and to guarantee dynamic balance of the robot. The methodology was evaluated by simulating the model of the biped robot together with a closed loop feedback linearization controller. The proposed planning method can be applied to other mobile robots control systems as a structured pattern generator which just needs initial and final kinematic information of the physical system at via points.