Adaptive consensus for a class of uncertain nonlinear multi-agent dynamical systems

This paper is concerned with adaptive consensus in multi-agent systems where each agent is dominated by an uncertain nonlinear dynamics, and adaptive controllers are presented for networks with and without time-delays, respectively. Consensus analysis is provided with Lyapunov functions for both cases, which shows that, with certain assumptions on the communication networks, consensus will be achieved asymptotically despite of parametric uncertainty in each agent. Without assuming that dynamics of each agent are passive as in the literature, the proposed control strategy adaptively renders the system passive. Moreover, this approach does not require linearity in the unknown parameter as assumed in our previous work. It can be obtained for undirected networks analogously, while the analysis here focuses on directed networks. © 2011 IFAC.