Data-Driven Stabilization for a Class of Unknown Switched Linear Systems with Asynchronism

This brief investigates the data-driven stabilization for a class of unknown switched linear systems in both continuous and discrete time. We assume that the switching signal is known but the system matrices of each subsystem are unknown and belong to some sets described by finite noisy input-state data. Data-driven parameterization of each unknown subsystem is built on the instantaneous bound of the external disturbance rather than the energy bound. The closed-loop stability is guaranteed by adopting a controller-mode-dependent Lyapunov function and the resulting stability conditions are data-based rather than model-based. The switching signals are arbitrary as long as they satisfy the average dwell time (ADT) condition that derived from noisy data. Two numerical examples are presented to illustrate how to enhance the feasibility of the proposed methods.