Fuzzy constrained predictive control of non-linear systems with packet dropouts

This study investigates the problem of fuzzy predictive control of non-linear systems with imperfect communication links. Packet loss (which appears typically in a network environment) is assumed to happen intermittently between the physical plant and the controller, and stochastic variables satisfying the Bernoulli random binary distribution are utilised to describe the imperfect communication phenomenon. The formulation is mathematically transformed into a stochastic Takagi-Sugeno fuzzy model. Attention is focused on the design of fuzzy predictive controllers such that the closed-loop system is stochastically stable, optimising an objective function value at every step in an infinite time horizon subject to input constraints and packet dropouts. A piecewise Lyapunov function approach is utilised, which is effective for fuzzy systems with trapezoidal membership functions; for continuous membership functions, the quadratic Lyapunov function approach is employed. A set of linear matrix inequalities is given to solve the corresponding controller optimisation problem. Two examples are provided to illustrate the usefulness and applicability of the developed theoretical results.