Adaptive neural consensus of nonlinearly parameterized multi-agent systems with periodic disturbances

Abstract

This article settles consensus of nonlinearly parameterized multi-agent systems with periodic disturbances by using matrix theory, adaptive control, neural networks and fourier series expansion. Firstly, uncertain nonlinear dynamics with unmeasurable periodic input disturbances are constructed and described by using fourier series expansion and neural networks. Secondly, a novel distributed control protocol based on adaptive control method and matrix theory is designed to make the second-order closed-loop systems asymptotically stable. Thirdly, another new distributed control protocol based on the above consensus protocol is designed to make the closed-loop system with unknown control directions asymptotically stable. Finally, the correctness of the two control protocols is verified by three simulation examples.