Heuristic Generation of Driving Input and Control of Brachiation Robot

This paper deals with a new type of dynamically mobile robot developed by modeling a gibbon, which is a long-armed ape that moves from branch to branch swinging its body like a pendulum. Since it makes use of gravity to swing, it does not use much force to move. The manner of moving like a gibbon is called brachiation. In this paper, we present a new method to control a brachiation robot and describe the results of simulations using a two-link-model robot. The main subject of this study is how to control it naturally and dynamically while taking advantage of gravity. First, we introduce a heuristic approach in which locomotive forces are generated by the robot itself after familiarizing repetitions. Next, we show that interpolation utilizing the spline function between heuristically generated driving inputs creates new forces that can accommodate the robot to previously unexperienced conditions. We also show that trajectory feedback control based on feedforward control achieves natural and dynamical locomotion of the brachiation robot.