Observer design for a class of hyperbolic PDE equation based on a Distributed Super Twisting Algorithm

Abstract

In this paper a new version of a Distributed Super-Twisting Algorithm (DSTA), including a linear term, is proposed. It is an extension to infinite dimensional spaces of the Generalized Super-Twisting Algorithm for finite dimensional systems proposed in [14], [15], [3]. The proposed algorithm is different from the one presented previously by [18], [22] and it retains all the main properties of its finite dimensional counterpart, that is, it converges in finite time to zero, even in presence of bounded perturbations, in contrast with the asymptotic convergence and weaker robustness properties that have been shown for the algorithm in [18], [22]. This properties are shown using a strong Lyapunov functional. As application of this algorithm the finite time and robust state estimation problem for a class of uncertain hyperbolic PDEs is considered. A numerical example illustrates the effectiveness of the proposed method.