HUMANOID ROBOT MOTION PLANNING – A MULTIPLE CONSTRAINS APPROACH

Abstract

In this paper, we present a new method for humanoid robot motion planning under multiple constraints. In our method, the multiple constraints humanoid robot motion is formulated as a multiobjective optimization problem, considering each constraint as a separate objective function. Three different constrains are considered: (1) minimum energy consumption; (2) stability; and (3) walking speed. The advantage of the proposed method is that in a single run of multiobjective evolution, are generated humanoid robot motions satisfying each constraint. The results show that optimal humanoid robot gaits have a large similarity with that of humans. In order to further verify the performance of optimal motions they are transferred to the “Bonten-Maru” humanoid robot.