Semi-passive Walking Control of Biped Robot with Muscle Model

In this paper, the plane stable walking control problem of a semi-passive biped robot with a muscle model was studied. The spring in the Spring-loaded inverted pendulum (SLIP) model was replaced by the modified Hill muscle model, and the dynamic equations of the biped robot in the single support and double support stages were established by the Lagrange method. Human-like control strategies include a combination of passive and active contraction of the leg muscles to keep the total energy of the system near a certain value and controlling the height and forward velocity of the particle by adjusting the landing position of the swinging leg. Based on theoretical analysis, simulation was carried out in the MATLAB environment. The results show that the control method proposed in this paper can realize the stable periodic walking of the biped robot based on the muscle model. When the system is adjusted with external disturbance, the total energy can be restored to balance and the robot can re-enter the stable periodic walking. Moreover, compared with the SLIP model, the proposed control method has stronger robustness.