Conseusus Research Of Multi-Agent Systems With External Disturbances

In recent years, more and more attention has been paid on cooperative control of multi-agent systems owning to its various applications in artificial intellegence, spaceflight, sensor networks, etc. The consensus problem of multi-agent systems, as one important cooperative control problem, is to design a distributed algorithm to make all agents reach an agreement on certain quantities of interest through limited information exchange between neighbors. This paper addresses the consensus problems of multi-agent systems with external disturbances. The main works of this paper are as follows:(1) For the following two cases, multi-agent systems with given model of the plant and plant dynamics unknown, we design two different kinds of nonlinear observers based on local output information. Based on velocity and disturbance estimate value, a distributed consensus algorithm is proposed for multi-agent systems with an undirected topology, and the influence of disturbance can be eliminated. Applications of this consensus algorithms to multi-vehicle cooperative control are also discussed.(2) Average consensus problem of muti-agent systems with general linear dynamics and external disturbances is considered. In the study of multi-agent systems with continuous-time general linear dynamics under undirected topology, we design a linear extended state observer based on local information, and a consensus algorithm is proposed based on local and neighbor information. It is shown that the consensus problem of high-dimensional multi-agent systems can be turned into the stability problem of a set of systems with the same dimension as one agent. In the study of multi-agent systems with discrete-time general linear dynamics, we design full-order and reduced-order observers and propose a consensus algorithm based on relative output information. By solving the modified algebraic Riccati equation, the controller and observer gain can be obtained.(3) This paper addresses the Leader-Follower consensus problem for multi-agent systems with identical general linear dynamics and unknown external disturbances. This paper addresses the Leader-Follower consensus problem for multi-agent systems with identical general linear dynamics and unknown external disturbances. Then a consensus algorithm is proposed for each agent based on the relative estimated states between neighbors and its own disturbance estimate. Finally, we present a simulation example to demonstrate the effectiveness of the proposed algorithm.