وحيد رضايي

During the past few decades, research in the field of cooperative control of multi-agent systems has attracted the attention of many researchers. Robotic manipulators, DC microgrids, unmanned vehicles, sensor networks, an‎d object transportation are just a few examples of a wide range of applications in which distributed cooperative control algorithms have been applied. The main goal of all these distributed algorithms is to design local distributed controllers for each of the agents in order to reach a final goal so that the agents are only allowed to exchange information with their neighbors. Depending on what the end goal is, various distributed algorithms such as formation, consensus, rendezvous, an‎d flocking have been developed for networked systems. Among these algorithms, the algorithm that has been given special attention in the literature of distributed systems is the algorithm related to the consensus of agents in a multi-agent system. Consensus in multi-agent systems refers to the event that agents, by sharing information with their neighbors, finally agree on a common value. In this thesis, a new protocol for the consensus control of second-order linear agents is proposed so that the new minimum-maximum criterion is used to bring the agents to a consensus. In fact, the proposed protocol produces a control set point for each agent at each stage of its execution, by finding the minimum an‎d maximum acceptable states that guarantee to achieve the goal of consensus. Then, by applying the control inputs, each agent of the network is driven to the control set point defined for it in each stage of the above protocol execution. In order to increase the speed of convergence, the proposed protocol considers different working modes called Tracker, Following an‎d Cross for network agents so that each agent can be in one of these three working modes. Unlike most of the existing works where the link weights of the communication graph must have a positive lower bound so that the convergence is realized asymptotically, in this work, the switching communication graph is assumed to be connected with undetermined weights. In fact, the proposed protocol can bring the agents to a consensus under more general conditions. In order to demonstrate the efficiency of the proposed protocol, numerical simulations are presented an‎d compared with conventional consensus protocols. Numerical simulations show that the proposed protocol has a significant convergence speed compared to conventional works.

كنترل مشاركتي , اجماع , چندعامله , توپولوژي سوئيچينگ , سرعت همگرايي

Cooperative control , Consensus , Multi-agent , Switching topology , Convergence speed

Vahid Rezaei

Esmaeel Khanmirza