Research On Synchronization Of Complex Networks And Consensus Of Multi-agent Systems

Recently, as an important approach for describing and understanding com-plex systems, complex networks and multi-agent systems have attracted many researchers from different disciplines. The research on the synchronization of the complex networks and the consensus of the multi-agent systems have been a more meaningful and more challenging object in control community field. Based on the former research, the thesis gives a survey on the synchronization of the gen-eral coupled complex networks and the complex networks consisted with delay nodes. Simultaneously, the consensus and flocking problem of the leader-follower multi-agent systems is also investigated on the basis of some related references on a theoretical and practical background.The main contents and contributions of this dissertation are summarized as follows:1. Pinning synchronization of the complex networks with non-delayed and delayed couplingIt is noticed that most of the studies on synchronization of dynamical net-work have been performed under some implicit assumptions that there exists the information communication of nodes via the edges only at time t or at time t-Ï„. However, in many circumstances, this simplification does not match satisfactorily the peculiarities of real networks:there exists the information communication of nodes not only at time t but also at time t-Ï„. Whereas the synchronization of both delay-coupled and non-delay coupled complex dynamical network have almost been ignored in the literatures. Motivated by this, we study the synchro-nization of the networks with both delayed and non-delayed coupling. We design controllers to ensure that the special networks could get synchronization. To re-duce the number of the controllers, we control the complex networks by pinning part of nodes.2. Pinning synchronization of complex networks with delayed nodesTime delays are common in many systems and time delays in the couplings have been much investigated in the literature. For example, time delays in the dynamical nodes, on the other hand, is a much more complex and important phenomenon that is still relatively unexplored with numerous real-world problems being modeled by delayed differential equations for which the delays occur at dynamical nodes. Therefore, this paper investigates the global synchronization of complex networks with delayed nodes by using pinning controllers. Leader-follower approach provides an effective route to consensus of the multi-agent systems.3. Consensus of multi-agent systems with an active leader and asymmetric adjacency matrixA neighbor-based observer was designed to estimate the unmeasurable state of an active leader in the multi-agent systems. Note that the interconnec-tion between agents of the multi-agent systems was described by the undirected graphs. That is, the adjacency matrix is symmetric. However, in many real-world the multi-agent systems, the interconnection between agents is more likely to be described by directed graphs. That is, the corresponding adjacency matrix is asymmetric. So we further explores the consensus of the multi-agent systems with an active leader and asymmetric adjacency matrix.4. Second-order tracking control for leader-follower multi-agent flocking in directed graphs with switching topologySecond-order dynamical models, in fact, are required for a broad class of agents in the real world, especially in problems where the agent dynamics can be feedback-linearized as double integrators (such as the holonomic mobile robot dynamic models). Therefore, besides first-order dynamics, second-order control protocols is also very important in consensus algorithm of the multi-agent sys-tems. There is hence a genuine need to consider the flocking problem for a class of leader-follower multi-agent systems using second-order tracking protocols in directed graphs with switching topology.