Distributed Observer-based Adaptive Consensus Protocols for Leader-following Multi-agent Systems

This paper considers the consensus problem of leader-following multi-agent systems under the undirected and connected interaction topology. The dynamics of agents is modeled by the general form of linear system. It is assumed that only a subset of the following agents can access to the state information of the leader, all following agents can not access to the bounded reference input of the leader, and the states of all following agents can not be available directly. To achieve control goal, local state observers are adopted by the agents to estimate its states, by which the distributed observer-based protocols are constructed to solve the leader-following consensus problems. To implement the protocol in fully distributed fashion, the parameters involved in the protocols are governed by the adaptive laws. The full-order and reduce-order observer-based adaptive consensus protocols can be unified in our proposed framework. To track the active leader, a discontinuous control protocol based on the state observers is proposed for each following agent. It is shown that under suitable conditions, all the following agents can track the leader under the provided protocols. Finally, a numerical example is given to illustrate our obtained result.