| In recent years, how to coordinate a group of spatially distributed dynamic systems or agents have become a hot area of research within the context of multi-agent systems. As a fundamental problem in the study of the distributed coordination for multi-agent systems, the consensus problem for multi-agent systems has attracted more and more attention recently. In a team of agents, "consensus" means that all the agents reach an agreement on their certain variables of common interest states, and consensus algorithm is a designed networked protocol for agents reach agreement based on local information exchange. An important control strategy among many others is the leader-following consensus among a group of agents.In this paper, we focus on investigating consensus problem for second-order multi-agent systems with a leader. First, we study the leader-following consensus problem for multi-agent systems with directed networks. We discuss the leader-following consensus problem for multi-agent systems when the corresponding directed graph of the network is fixed and not necessarily strongly connected. The designed distributed coordination laws for the followers adopts the agents'velocity information, then based on the classical Lyapunov theory, the stability of the closed-loop system is analyze in detail. The effect of the aggregation gains on the tracking performance of the closed-loop system is theoretical proved.In practical networked multi-agent systems, due to the effect of random communicat ion environment, noises are unavoidable during information acquisition and transmission among agents. Then the leader-following consensus problem with measurement noises and directed network topology is further investigated. We propose a consensus protocol which can attenuate the effect of measurement noises. By utilizing ito stochastic differential analysis method, a sufficient condition for second-order multi-agent system achieving consensus in mean square sense is obtained. |
PDF Full Text Request