Modeling and Control of Passive Dynamic Walking Robot with Humanoid Gait

Considering the sagittal movement and the lateral swing in the humanoid practical walking, a new humanoid passive dynamic bipedal robot with the lateral movable upper body is proposed in this paper. The finite state machine (FSM) theory is adopted to control the robot, which changes agilely the control strategy according to the practical states of the humanoid gait. In the method, the torque compensation adaptive excitation control strategy is used for sagittal control and PID is applied to the upper body for the robot's lateral stability. It is verified by the co-simulation using ADAMS and MATLAB that the bipedal robot can reach the stable humanoid gait with the high energy efficiency.