Enlarging Domain of Attraction of Robust Predictive Control for Constrained Piecewise Affine Systems

CHEN Fu,ZHAO Guang-zhou
DOI: https://doi.org/10.3785/j.issn.1008-973x.2009.11.012
2009-01-01
Abstract:Aimed at a class of constrained discrete-time piecewise affine(PWA)systems with addictive bounded disturbances,a new method for enlarging the domain of attraction of robust model predictive control(MPC)was presented.The maximal robust positively invariant set and the associated local stable control law were computed.Based on the maximal robust positively invariant set,a contractive sequence of robust stabilizable sets were computed off-line using multi--parametric programming with one-step predictive horizon.By considering this sequence as terminal constraint of predictive states in the optimization problem,the resulting controller can robustly steer the states in the contractive sequence of sets to the maximal robust positively invariant set.Hence the stability and the enlargement of domain of attraction of robust predictive control were guaranteed.A numerical example showed the validity of the proposed method.
What problem does this paper attempt to address?