A Time-Varying Observer-Based Approach to Equilibrium Estimation and Compensation for Synchronization of Heterogeneous Nonlinear Cyber-Physical Systems

This study focuses on the time-varying observer design for the synchronization of heterogeneous nonlinear cyber-physical systems, where each subsystem can be modeled by an agent, and the information flow among the subsystems can be regarded as the interactions among the agents. To achieve the synchronization behaviors, a novel estimation and compensation mechanism based on time-varying observers, for the first time, is proposed. The mechanism can reveal the influence of equilibrium points on the synchronization behaviors of a class of heterogeneous systems with nonlinear dynamics and unknown external inputs. Specifically, time-varying observers are utilized to construct the proper estimation and compensation controllers using the limited and dynamic information interaction. By developing a novel mechanism of the dwell time of switching communication topologies, it is proven that the stability conditions of perfect synchronization can be achieved for the closed-loop heterogeneous nonlinear systems with external unknown inputs. The effectiveness of the developed time-varying observers in the synchronization of motion estimation applications is demonstrated by heterogeneous robotic systems.