Energy-efficient reference gait generation utilizing variable ZMP and vertical hip motion based on inverted pendulum model for biped robots

Abstract

Energy-efficient reference gait generation algorithm is suggested utilizing variable ZMP (Zero Moment Point) and vertical hip motion, which maxmizes the energy efficiency for cyclic gaits, based on inverted pendulum model (IPM) for biped robots. As observed in natural human walking, the robot's vertical hip motion (with average height and amplitude of hip ripple) as well as allowing variable ZMP pattern, which reduces the joint stress (i.e., rapid acceleration and deceleration of body) to minimize a suitable energy performance - amount of energy required to carry a unit weight a unit distance. In order to ensure physically realizable walking trajectory, we also consider joint velocity limits, kinematic constraints, friction constraint, and yawing moment constraint. Simulations are performed with a 12-DOF 3D biped robot model to verify the effectiveness of the proposed method.