Observer-Based Optimized Time-Varying Formation Control for Second-Order Multi-agent Systems Under Directed Networks

In this paper, the observer-based optimized time-varying formation control problem is investigated for a class of second-order multi-agent systems under directed networks. Considering that not all agents have access to the virtual leader, a distributed finite-time observer is developed for each agent to obtain the target states, where the global information is not required. By utilizing observed information, a distributed optimized control strategy based on the improved reinforcement learning algorithm is presented to tackle the time-varying formation control problem. Simulation results are given to verify the effectiveness of the proposed method.