Balance Control For Humanoid Robot Biped Walking

Humanoid robot is one of the most active research area of intelligent robot. The research of humanoid robot aims to facilitate human's life. However the human environment is complex and dynamic. In order to apply the robot in human's living environment,its stable walking is necessary. Therefore, the research of balance control for humanoid robot biped walking is significant in both the theory and practice.This paper firstly introduces the research and development status of humanoid robot technology. Then, the problem of biped walking is studied, and corresponding control strategies are proposed to solve landing impact force and walking stability. Finally, the balance control strategies of robot walking are analyzed and summarized, and some improvements are proposed. The main work of this paper includes:Compliance control is realized to absorb the landing impact force. According to the force/torque information of F/T sensor, the movement of swing leg is controlled to reduce the landing force.Stability monitoring of robot is established. The foot trajectory is calculated by forward kinematics, and the force/torque information can be acquired by sensor, so the walking ZMP is calculated online to monitor the stable status of robot.ZMP compensation method is designed. By modifying the trajectory of the hip center, the deviation between the actual position of ZMP and the reference input is compensated to enhance the stability of dynamic walkingThe algorithm of model predictive control with state estimation is studied and applied in balance control of robot. The trunk model is used to predict the output of robot, and with the feedback correction and rolling optimization, the disturbance is corrected and the actual ZMP position followed the reference input.