Event-based Consensus Problems Of Networked Multi-agent Systems

In recent years, synthesis and analysis of multi-agent systems have been extensively investigated by various researchers from multiple areas. As a fundamental issue in distributed cooperative control of multi-agent systems, the consensus problem means that all agents using consensus protocols eventually reach a common value on certain quantities of interest. In order to achieve consensus, each agent needs to updates its control input by exchanging local information with its neighbors. In the practical application, an agent may be equipped with an embedded microprocessor to collect and compute information. However, the power supply of an agent is always limited and most existing consensus protocols require continuous communication and measurements. From that respect, it is necessary for researchers to devise a proper consensus algorithm with lower energy consumption that is event-triggered consensus control protocol. Applying this method, all agents need to execute control updates only at some specific instants on the premise that systems achieve consensus. Therefore, the energy consumption can be further reduced and the practicability can be improved.This paper investigates the event-triggered consensus problems of multi-agent systems with and without a virtual leader, obtaining the following results:(1) We study the event-triggered consensus problems of first-order multi-agent systems without a virtual leader. Considering the power consumption of communication subsystems, we propose a novel event-triggered sampled-data consensus protocol. The algebra graph theory, Lyapunov stability method and some other methods are employed to acquire the convergence condition gauranteeing that first-order multi-agent systems with an undirected connected network topology reach average consensus. Furthermore, numerical simulations show that compared with the existing protocol, the proposed protocol can further reduce the energy consumption.(2) We consider the event-triggered consensus tracking problems of first-order multi-agent systems with a virtual leader. By introducing the sampled-data and tracking error into the event condition design, we propose a new event-triggered consensus tracking protocol. The algebra graph theory, Lyapunov stability method and some other techniques are used to obtain the convergence condition gauranteeing first-order multi-agent systems with an undirected connected network topology and a virtual leader to achieve consensus tracking. Moreover, numerical simulations show that compared with the protocol without sampled-data, the proposed protocol can further save the energy consumption.(3) We investigate the event-triggered consensus tracking problems of second-order multi-agent systems with a virtual leader. By adding the sampled-data and tracking error into the event condition design, we propose a new event-triggered consensus tracking protocol. The algebra graph theory, Lyapunov stability method and some other techniques are used to obtain the convergence condition gauranteeing that second-order multi-agent systems with an undirected connected network topology and a virtual leader achieve consensus tracking. Moreover, numerical simulations show that compared with the protocol without sampled-data, the proposed protocol can further save the energy consumption.