Multiobjective locomotion optimization of quadruped robot with different 2DOF configurations of actuated spine

Abstract

This paper addresses the problem of finding a set of optimal foot and spine trajectories for a biologically inspired quadruped robot in order to obtain energy optimal and fast locomotion in forward direction. Based on the previous work of finding optimal foot trajectories we now extend the scope of problem by investigating the effects of an added 2DOF actuated spine on the characteristics of robot locomotion in comparison with rigid spine. To tackle this problem a multiobjective genetic optimization algorithm was used. Purpose of the optimization procedure is to give a better insight in which of the two spine configurations is more beneficial to studied bound and trot gaits and how to design a control strategy that utilizes the spine in an effective way.