Design and Validation of a Model-Based Automated Controller Tuning Algorithm

This paper presents a model-based automated controller tuning algorithm for repetitive references. A key challenge in automated tuning is guaranteeing stability of the closed loop system during the tuning process. In order to guarantee the system stability while tuning, we reformulate the traditional controller tuning problem into the tuning of the Youla parameterized version of the controller. Thus at each iteration of the tuning process, the Youla parameter is iteratively tuned to minimize a given quadratic cost function. This makes the optimization problem affine in the parameters and at the same time, eliminates the need for checking stability at each iteration. The proposed algorithm can be used for (1) auto-tuning for optimizing a known controller, or (2) auto-tuning a controller from some arbitrary controller candidates. These capabilities of the model-based tuning algorithm are finally demonstrated experimentally on a precision linear stage. Criteria for convergence of the tuning law has also been presented.