The structure of the controllable set for multimodal systems (Q1098804)

Multimodal systems arise naturally in the study of multirate sampled-data systems [see \textit{D. P. Stanford} and \textit{L. T. Conner}, SIAM J. Control Optimization 18, 488-497 (1980; Zbl 0454.93008)] and have applications in switched capacitor circuits. In the paper under review the authors investigate the structure of the controllable set of a multimodal system. They define a maximal component of the controllable set, investigate the controllable set as the union of its maximal components and show that for each positive integer k, state dimension \(n\geq 3\), and control dimension \(m\leq n-1\) there is a multimodal system whose controllable set S(L) is the union of exactly k maximal subspaces of \({\mathfrak R}^ n\), and this system has k as a bound on the number of iterations necessary to reach any state in S(L) from zero. They also prove that the above holds for \(k=\infty\) and show that for each state dimension n and each control dimension m, there is a completely controllable multimodal system having bound \(2n-2m\).