Cooperative control of multi-agent systems has been a hot topic in the control and robotics communities in the past two decades, due to many practical applications such as formation control of mobile robots, coordination of unmanned aerial vehicles, and coverage control of sensor networks. Various cooperative control problems including but not limited to consensus, synchronization, formation, and cooperative output regulation, have been extensively investigated. These cooperative control problems are often addressed via distributed control strategies, in which each agent can only use the information of itself and its neighboring agents.

In practice, individual agents may have limited resources such as energy resources and communication bandwidth. To improve their utilization efficiency, many distributed event-triggered control strategies have been developed for multi-agent systems. However, most of them share a fundamental but undesirable need to use some global information on the communication graph and/or continuous communication between neighboring agents.

This thesis will investigate some cooperative control problems of multi-agent systems via event-triggered control by overcoming the aforementioned two limitations. The main objective is to develop fully distributed event-triggered control strategies composed of an event-triggered control law and an event-triggering mechanism, such that the cooperative control problem is solved with intermittent communication. This thesis focuses on consensus and cooperative output regulation problems, and can be divided into two parts. The first part addresses the leaderless consensus problem of linear multi-agent systems under fixed topologies via fully distributed event-triggered control strategies. The main results of this part are summarized as follows:

1. The state consensus problem is studied for homogeneous linear multi-agent systems subject to external disturbances. A distributed event-triggered adaptive output feedback control strategy is proposed for each agent. To avoid continuously monitoring the neighboring information, a distributed self-triggered adaptive output feedback control strategy is further designed. Under the proposed two control strategies, the consensus problem can be solved in a fully distributed manner.

2. The output consensus problem is investigated for heterogeneous linear multi-agent systems. By constructing a dynamic compensator, a fully distributed event-triggered control strategy is proposed for each agent. Under the proposed control strategy, all agents asymptotically achieve output consensus with intermittent communication in a fully distributed manner.

The second part of this thesis addresses the cooperative output regulation problem of heterogeneous linear multi-agent systems under fixed topologies via fully distributed event-triggered control strategies. This problem is more general than the leader-following consensus problem because it aims to make all agents achieve not only reference tracking but also disturbance rejection. The main results of this part are summarized as follows:

1. The cooperative output regulation problem is investigated, where each agent knows the system matrix of the exosystem. By using the feedforward design approach, a fully distributed event-triggered dynamic output feedback control law is proposed. Meanwhile, a fully distributed dynamic event-triggering mechanism is designed so that each agent can determine when to broadcast its information to its neighbors. With the proposed control strategy, the cooperative output regulation problem can be solved in a fully distributed manner by intermittent communication.

2. The cooperative output regulation problem is studied, where only the agents connected to the exosystem know the system matrix of the exosystem. To estimate both the state and the system matrix of the exosystem, an event-triggered adaptive distributed observer is designed for each agent. Then, based on this distributed observer, a fully distributed event-triggered dynamic output feedback control law and a fully distributed event-triggering mechanism are proposed for each agent. With the proposed control strategy, the cooperative output regulation problem can be solved in a fully distributed manner by intermittent communication under a mild condition.

3. The cooperative output regulation problem is investigated, where the agents contain uncertain parameters and the exosystem contains unknown parameters. By utilizing the internal model principle and the adaptive control technique, a distributed adaptive internal model is constructed for each agent. Then, based on this internal model, a distributed event-triggered adaptive output feedback control law and a distributed event-triggering mechanism are proposed. With the proposed control strategy, the cooperative robust output regulation problem can be solved without requiring either the global information on the communication graph or the bounds of the uncertain or unknown parameters. The special case with a single agent is also considered.

In the above two parts, the feasibility of the proposed event-triggering mechanisms is demonstrated by strictly excluding Zeno behavior for each agent.