Data driven predictive control based on orthonormal basis functions

This paper presents a concept of an adaptive model predictive control (MPC) scheme based on a flexible predictor model that utilizes orthonormal basis functions (OBFs). This model structure offers a trade-off between adaptation of the model accuracy in terms of the expansion coefficients and the dynamical structure in terms of the basis functions. We show that this adaptation can maintain desirable control performance. Moreover, since OBF model structures can be seen as a generalization of finite impulse response (FIR) model structures, the incorporation of this scheme in FIR-based MPC is rather straightforward.