Decentralized Adaptive Control for A Class of Nonlinear Time-Varying Interconnected Systems

In this paper, we propose decentralized adaptive control for a class of nonlinear time-varying interconnected systems. The overall system may be represented as a set of decentralized linear time-varying subsystems with time-varying nonlinear interconnections. We first analyze the effect of the time variations of parameters and interactions for the adaptive system and show that the proposed decentralized adaptive system is ultimately bounded with respect to a residual set. Then under the assumption that the structure of the linear time-varying subsystem is known a priori, we show that the proposed decentralized adaptive control scheme is capable of tracking a local reference trajectory for each subsystem asymptotically with a smaller residual set. A simulation of two identical pendulums which are coupled by a moving spring is presented to validate our conclusions.