Hybrid Persistency of Excitation in Adaptive Estimation for Hybrid Systems

TL;DR

This paper introduces a hybrid framework extending persistency of excitation concepts to hybrid systems, establishing conditions for stability and demonstrating advantages over traditional methods in adaptive estimation scenarios.

Contribution

It generalizes PE and UO notions to hybrid systems, linking them to stability, and applies these results to adaptive estimation problems with improved performance.

Findings

Hybrid PE implies hybrid UO, leading to stability.

The framework outperforms classical continuous/discrete methods.

Applications include adaptive observers for nonlinear hybrid systems.

Abstract

We propose a framework to analyze stability for a class of linear non-autonomous hybrid systems, where the continuous evolution of solutions is governed by an ordinary differential equation and the instantaneous changes are governed by a difference equation. Furthermore, the jumps are triggered by the influence of an external hybrid signal. The proposed framework builds upon a generalization of the well-known persistency of excitation (PE) and uniform observability (UO) notions to the realm of hybrid systems. That is, we establish conditions, under which, hybrid PE implies hybrid UO and, in turn, uniform exponential stability (UES) and input-to-state stability (ISS). Our proofs rely on an original statement for hybrid systems, expressed in terms of Lp bounds on the solutions. We demonstrate the utility of our results on generic adaptive estimation problems. The first one concerns the so-called gradient systems, reminiscent of the popular gradient-descent algorithm. The second one pertains to designing adaptive observers/identifiers for a class of hybrid systems that are nonlinear in the input and the output, and linear in the unknown parameters. In both cases, we illustrate through examples that the proposed hybrid framework succeeds when the classic purely continuous- or discrete-time counterparts fail.