Balance control strategy of humanoid robot based on stiffness control

Abstract

For the control system of humanoid robot, a fundamental function should be able to keep its balance under a certain disturbance, which is also one of the most important topics in the study of bipedal robotics. In this paper, a novel scheme of the ankle and hip balance control strategies for humanoid robot stability are proposed. Firstly, the external disturbance force is estimated from kinematics of robot by using angular momentum theorem. For small disturbance force, the angular acceleration for ankle rotation is calculated using inverted pendulum model combining with ZMP criterion. By controlling the stiffness of DC motor, the ankle strategy is achieved. For large disturbance force, the maximum stable region is deduced using flywheel inverted pendulum model combining with ZMP criterion, then the relationship for balance state between angular velocities of ankle and hip is established by using reaction null-space theory. Moreover, to recovery initial status of robot, PD feedback control method is applied on both strategies. Experimental results with a NAO humanoid robot indicate that the proposed scheme for balance control is correct and effective..