融合前馈与姿态预测的并联稳定平台自抗扰控制策略

This research proposes an active disturbance rejection control (ADRC) method with a feed-forward compensator for a parallel stability platform based on its attitude prediction model. The time delay of the control system responding to high frequency inputs and the nonlinear disturbances arising from loading conditions and frictions are considered to improve the tracking performance. Through analyzing the dynamics of the electric cylinders and the kinematics of the parallel mechanism, the control model in state space together with the kinematic inverse model of the system are introduced. An extended state observer (ESO) is designed to observe the nonlinear disturbances of friction behaviors and loading conditions, and the non-linear tracking differentiator (TD) is used to realize a rapidly response of the electric cylinders without overspeed alarm. With the observation results and the TD, a closed-loop controller with a feed-forward compensator based on the attitude prediction model of the platform is presented to improve tracking accuracy and response speed. The control method is verified on a 3 degree-of-freedom (DOF) parallel stability platform, and comparative experimental results indicate that under different loading conditions, the proposed controller can achieve satisfactory tracking performance for the 3-DOF platform with an increased response speed and a reduced tracking error.