Data-Driven Iterative Cooperative Tuning of Feedback and Feedforward Controllers for Precision Motion Systems

Existing research on precision motion control methodologies typically focuses on separate design of feedback control and feedforward control. As trajectory tracking accuracy requirements for precision motion systems become increasingly stringent, individually optimizing a single controller faces growing challenges in meeting performance demands; moreover, the sequential design process for feedback and feedforward controllers is lack of efficiency. To address these limitations, this paper proposes an iterative parameter tuning method that employs an efficient data-driven framework to achieve cooperative optimization of feedback and feedforward controllers. Different from model-based controller design approaches, the proposed method utilizes trajectory tracking data to iteratively tune the controllers’ parameters via gradient-based optimization. The key step of the proposed method is to estimate the gradient and Hessian matrix of the cost function without prior knowledge of the system dynamics. Comparative experiments on an industrial precision motion stage demonstrate that the proposed method achieves significant trajectory tracking enhancement: peak error reduction from 6.28 × 10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−5</sup> m to 1.95×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−7</sup> m.