Two-player zero-sum games for leader-follower consensus of linear multi-agent systems with unknown dynamics

摘要

In this paper, the leader-follower consensus problem of linear multi-agent systems with unknown dynamics and external disturbances is studied by using the two-player zero-sum game and adaptive dynamic programming. First, we establish the global error dynamic system of the multi-agent system and decouple the global error dynamic system for each agent. Second, we use the two-player zero-sum game theory to deal with external disturbances, and get the generalized algebraic Riccati equation for each agent. Third, based on adaptive dynamic programming, we propose a new algorithm for solving the leader-follower consensus problem for multi-agent systems with unknown dynamics. At last, Simulation results are further presented to show the effectiveness of the proposed algorithm.