Stability analysis of nonlinear networked control systems with asynchronous communication : a small-gain approach

In this paper, we study the stability of decentralized networked control systems (NCSs) in which the sensors, controllers and actuators communicate through a finite number of local networks. These local networks accommodate the communication between local (decentralized) controllers at uncertain transmission times and operate asynchronously and independently of each other. In addition, each of the local networks exhibits communication constraints that require the presence of a protocol that decides which of the (local) network nodes is allowed to transmit its corresponding information at which transmission time. Due to the asynchronous nature of the networks, most existing works on the stability analysis of NCSs are not applicable as their stability characterizations assume that there is only one global communication network, or at least one global coordinator (or clock). Therefore, we present a novel approach that leads to maximal allowable transmission intervals for each of the individual local networks that guarantee the global asymptotic stability of the overall closed-loop system. The approach combines ideas from emulation-based stability analysis for NCSs and techniques from the stability of largescale systems.