Infinite horizon sparse optimal control

From MaRDI portal
Publication:2359778



Abstract: A class of infinite horizon optimal control problems involving Lp-type cost functionals with 0<pleq1 is discussed. The existence of optimal controls is studied for both the convex case with p=1 and the nonconvex case with 0<p<1, and the sparsity structure of the optimal controls promoted by the Lp-type penalties is analyzed. A dynamic programming approach is proposed to numerically approximate the corresponding sparse optimal controllers.


A class of infinite horizon optimal control problems with \(L^p\) cost functionals \((0<p\leq1)\) in the controls is studied. The existence of an optimal control for the full nonlinear dynamics is proved for the convex case \(p=1\), while for the nonconvex case \(0<p<1\) the existence is proved for the special case of affine controls. The sparsity structure of optimal control, favored by the nonsmooth cost functionals, is subsequently analyzed using first order optimality conditions. The presented concepts are illustrated with an analytical example of Eikonal dynamics with \(L^1\) cost, and several numerical examples involving the same system as well as a nonlinear system for both the convex and nonconvex cases are given.




Cited in
(26)








This page was built for publication: Infinite horizon sparse optimal control

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2359778)