Consensus decentralized control of multi-agent networked systems using vector Lyapunov functions

Abstract

Consensus control problem for a group of agents in a multi-agent networked system is considered. The agents are modeled by linear dynamical systems and the interconnection topology of agent's network is represented as a directed weighted graph. A decentralized approach to the synthesis of consensus control is proposed, including a synthesis of the local stabilizing controllers for each agent of networked system and synthesis of consensus controllers to ensure the conditions of agent's coordination. Local regulators design for individual agents is based on the method of invariant ellipsoids and reduced to solving a system of linear matrix inequalities. The stability of the multi-agent system taking into account the relationship between the agents is performed by vector Lyapunov functions method which leads to the stability analysis of aggregated comparison system with low dimension.