Design and Optimization of a Humanoid Bipedal Robot Based on Joint Coordinated Control

A bipedal robot design and control method based on joint coordinated control is proposed to improve the motion stability and operational efficiency of the bipedal robot. Firstly, time-based and event-triggered detection methods are combined to achieve stable ground contact detection, and the bipedal robot state is accurately estimated by fusing joint encoder and IMU (inertial measurement unit) data using a Kalman filter. Simulation results show that, compared to single time-based detection methods, the proposed approach exhibits better stability and achieves position estimation errors of less than 1% within a 10 m range. Furthermore, a joint coordinated controller based on a progressive gait control mechanism and torso tilt control mechanism is designed, significantly enhancing the walking stability. The progressive gait and torso tilt control strategies enable the robot to smoothly transit from standing to walking within 5 s. Through simulation, a cooperatively driven hip joint is also designed, with results demonstrating a 40% reduction in the maximum torque requirement for the hip yaw motor. Finally, a complete physical bipedal robot system is built. Experimental tests demonstrate the effectiveness of the proposed approach and its potential for real-world application.