Estimation based formation control with size scaling for leader-follower networks

This paper studies a multi-agent formation control problem under a leader-follower framework, where the agents are governed by double-integrator dynamics and the objective is to achieve a formation with desired shape and a specified size. Firstly, a distributed control algorithm is developed for the leaders to achieve a desired distance. Then, by combining the control law for the leaders and a relative position estimation based control law for the followers, an estimation based formation control algorithm is developed for the entire multi-agent system to asymptotically achieve a formation with desired shape and a specified size in the case without initial relative position estimation errors and approximately attain the desired formation in the case with initial relative position initial estimation errors. Simulation results are provided to validate the proposed algorithm.