Study On Control Strategy Of Humanoid Robotic Arm Joint Drive System

With the advancement of technology,the continuous development of computer technology,the continuous improvement of the level of automation,the rapid development of robot technology,more and more occasions need to use humanoid robots to work.The manipulative performance of the humanoid robot has an important impact on the actual working effect.Especially in the special industry of high precision requirements such as military and medical,the requirements for the stability and accuracy of the robot arm are extremely high.In order to improve the manipulative performance of the humanoid robot arm,this paper starts with the robot arm joint motor drive system,and studies the saturation non-linearity of the motor drive system and the torque ripple suppression of the arm drive motor.Aiming at the actuator saturation problem of robot arm joint drive system,the traditional anti-windup scheme is analyzed.Based on the traditional anti-windup controller,an anti-windup control strategy of humanoid robot arm joint based on integral state prediction is adopted.And the feasibility of the scheme is analyzed theoretically.By establishing the mathematical model of the arm joint of humanoid robot and combining with the analysis of the influence of speed controller model and integral initial value on the system,an integral state predictor is set up on the basis of the traditional anti-windup controller.Under the given conditions of different load torques and speeds,the final integral steady-state value of the integral link is estimated.When the integrator reaches saturation,the system stopped accumulating the rotational speed deviation.The predicted integral steady-state value at saturation is taken as the initial integral value of the speed controller entering the linear region,so that the system can quit saturation quickly.Torque ripple exists in the working process of the joint drive motor of the humanoid robot arm,and the amplitude of the torque ripple is larger in the commutation process.Aiming at the commutation torque ripple phenomenon of the arm joint drive motor,this paper adopts the model-free predictive control strategy based on the universal model and predictive control in the model-free control to suppress the torque ripple generated by the commutation process of the drive motor.By deducing the universal model equation of the arm joint drive motor control system and using it as the prediction model,the optimal switching state is selected by the value function to maintain the constant phase current value of the non-commutation winding of the drive motor,thus achieving the effect of maintaining the constant output electromagnetic torque,realizing the suppression of the torque ripple of the drive motor and improving the operation stability of the arm.The Matlab/Simulink software was used to build the simulation model to simulate the humanoid robot arm joint drive system.The simulation results show that the anti-windup control strategy of the human joint robot arm joint based on the integral state prediction can effectively exert the anti-windup effect,achieve the effect of suppressing the system overshoot,and ensure the system has good dynamic performance.The simulation proves that the model-free predictive control method can effectively suppress the torque ripple phenomenon in the commutation process of the drive motor,reduce the output torque ripple of the joint drive motor,and ensure the good operation performance of the arm.