Adaptive identification of continuous-time switched linear and piecewise linear systems

This paper presents a novel procedure for the online identification of continuous-time switched linear systems. The proposed procedure is an extension of the well studied series-parallel parameter identifiers in adaptive control. With the proposed results, it is possible to identify the submodels of a switched system online, independent of the nature of the switching signal. As a special case, the procedure is applied to the identification of piecewise linear systems with known regions. Analytical proofs for stability and parameter convergence are given. Numerical simulations validate the results.