Consensus In Complex Dynamic Network Of Multi-Agent Based On LMI Method

In this thesis, consensus problem of multi-agent networks is investigated by using linear matrix inequality (LMI) techniques. Based on Lyapunov stability theory combined with LMI, consensus problem of multi-agent networks can be solved via taking different network topology, time-delays among agents and external disturbances into account. The whole thesis consists of the following four parts:1. According to the meaning of transmitting information among agents, the different complex dynamic network topology could be established. Especially, in the multiple time-varying delays network, we study the average consensus problem for continune time undirected networks of multi-agent, coupling multiple time-varying delays by using linear matrix inequality (LMI) techniques.2. In the analysis processing of system's stablity, Lyapunov-Krasovskii function is constructured appropriately and some free weighted matrices are emplyed. Average consensus problem of multi-agent networks with multiple time-varying delays is consided by using linear matrix inequality (LMI) techniques. Numerical examples and simulations show the effectiveness and improvement of previous existng result.3. Average consensus problem for directed networks of multi-agent with both switching topology and time-varying delays is studied by using linear matrix inequality (LMI) techniques. Some free weighted matrices are emplyed in the analysis processing, which is less conservative and more general than the existing result.4. In reality, some variables of the agents in multi-agent network systems may not be measured precisely duo to external disturbances and time-delays. For different network topology (fixed topology and switching topology), we investigate consensus control with a time-varying reference state in directed multi-agent networks with time-varying delays and subject to external disturbances by using linear matrix inequality (LMI) techniques.