Support of COG trajectory tracking by arm-swing with bipedal walking

Walking ability, which is mainly composed of mechanical stability and moving efficiency, is one of the most important issues in the field of humanoid robots. An effective use of the robot´ s arms is expected to improve the walking ability under limited body constraints of the robot. We propose an arm-swing method to enhance the efficiency and the stability depending upon situations demands, which selects a weight of the COG trajectory tracking support. To support the COG trajectory tracking, we employ a predictive control of an inverted pendulum model by inertial force of arm-swing. Input weights and output weights in the predictive control are adjustable by the support weight selected by a selection algorithm for locomotion (SAL). The selection of the support weight (Su-SAL) keeps the stability maximizing the efficiency in comparison with the case that the support weight is constant.