Robust Gain-Scheduled PSD Controller Design from Educational Perspective

Paper in proceeding, 2016

Nowadays, the demand for qualified engineers in novel model-based controller design techniques is rapidly increasing. Therefore, it is necessary to include these up-to-date techniques in control theory courses. We are confident that this paper provides a motivation for embedding new linear/bilinear matrix inequality (LMI/BMI) -based approaches in topic-related courses and student projects. Within this paper a systematic design framework for robust discrete-time gain-scheduled proportional, summation, and difference (PSD) controller design for uncertain linear parameter-varying systems is presented. Conditions to satisfy robust stability and performance requirements are translated to an optimization problem subject to LMI/BMI constraints which students can easily solve using commercial as well as free and open source software tools and solvers. Two student project examples are given to illustrate and validate the proposed methodology.