Luca Zaccarian

We propose a family of nonlinear state feedback global stabilizers for all planar linear systems which are globally stabilizable by bounded inputs (namely, all non exponentially unstable linear systems). This family is parametrized by a nonlinear function whose selection can yield quasi time-optimal responses, where the ¿quasi¿ is required to achieve local exponential stability of the closed loop. The arising trajectories are quasi-time-optimal for arbitrarily large initial conditions; so, we expect the very simple proposed nonlinear control law to be very useful for embedded control applications with strong computational constraints.

In this paper we provide algorithms based on Linear Matrix Inequalities for the design of full-authority and external linear anti-windup compensators of reduced order guaranteeing finite global gain. Previous results showed that the reduced order anti-windup design problem is non-convex. We propose here a convex approximation of the non-convex constraints leading to constructive design algorithms. The proposed algorithms are successfully tested on two simulation examples taken from the literature.

In this paper we propose LMI methods for designing dynamic/static anti-windup compensators to improve regional performance and stability of linear control systems with saturating actuators. Algorithms are developed for minimizing the upper bound on the regional /spl Lscr//sub 2/ gain for exogenous inputs with /spl Lscr//sub 2/ norm bounded by a given value, and for minimizing this upper bound with a guaranteed reachable set or domain of attraction. Based on the structure of the optimization problems, it is shown that for systems whose plants have poles in the closed left-half plane, plant order dynamic anti-windup can achieve semiglobal stability and finite /spl Lscr//sub 2/ gain for exogenous inputs with /spl Lscr//sub 2/ norm hounded by any finite value. The problems are studied in a general setting where the only requirement on the linear control system is well-posedness and internal stability.

It is shown that uniform global exponential stability of the input-free discrete-time model of a globally Lipschitz sampled-data time-varying nonlinear system with inputs implies finite gain /spl Lscr//sub p/ stability of the sampled-data system for all p /spl isin/ [1, /spl infin/]. This result generalizes results on /spl Lscr//sub p/ stability of sampled-data linear systems and it is an important tool for analysis of robustness of sampled-data nonlinear systems with inputs.