Robust Model Predictive Control with Almost Zero Online Computation

This paper provides a strategy for the problem of robust model predictive control constrained, discrete-time systems with state and output disturbances. Using linear matrix inequality (LMI) method, nested geometric proportion asymptotically stable ellipsoid (GPASE) is designed off-line, then shrinking ellipsoids assures system converges on equivalent an exponential convergence velocity. The biggest advantage this method online computation almost reduced to zero, which makes it possible apply scheme not only plants slowly varying parameters, but also fast ones. Finally, simulation example shows validity proposed technique.