Neural-Network-Based Distributed Interval Observer Design for Nonlinear Multi-Agent Systems

A method of distributed interval observer for nonlinear multi-agent systems (MASs) using neural networks is studied in this paper. To address the issue that the traditional observer design methods can not be applied for unknown nonlinear functions effectively, radial basis function neural networks (RBFNN) are employed to estimate the upper and lower bounds of the unknown nonlinear term. For the error system, a Lyapunov function is chosen, which has the suitable network weight correction and network error selection mechanism, then the boundedness and non-negativity of the error dynamic system are guaranteed. Finally, a numerical example is used to demonstrate the effectiveness of the distributed interval observer.