Multi-rate self-tuning predictive controller for multi-variable systems

Abstract

A self-tuning control algorithm is presented for discrete-time, linear, multi-variable systems with skew output sampling and input updating mechanisms. In particular, it is assumed that the output components can be measured and the input variables can be updated at different times and rates. The control synthesis procedure relies on the predictive approach and is worked out by minimizing a suitably defined performance index. The presence of integral action in the control loop guarantees robust steady-state zero error regulation for constant reference signals irrespective of the sampling process adopted. Some simulation examples show the potential and limitations of the proposed approach.