Optimal stabilization of unstable periodic orbits embedded in chaotic systems

Abstract

A gradient-flow-based approach is proposed in this paper for stabilizing unstable periodic orbits (UPO) embedded in chaotic systems. In order to obtain an on-line stabilizing solution, the stabilization problem is considered to be an optimal control problem, and system state sensitivities with respect to the control input are introduced. The resulting feedback controller is able to stabilize UPO embedded in both kind of systems, with or without an odd Floquet number. Moreover, the proposed approach is easily extended to identifying the period of the UPO to be stabilized when it is unknown. Simulation experiments of the proposed controller are carried out on the Rössler and the Lorenz systems.