High-performance decentralized control for formation flying with leader-follower structure

In this paper, an incrementally linear decentralized control law is proposed for the formation of cooperative vehicles with leader-follower topology. It is assumed that each vehicle knows the modeling parameters of other vehicles with uncertainty as well as the expected values of their initial states. A decentralized control law is proposed, which aims to perform as close as possible to a centralized LQR controller. The behavior of the proposed decentralized strategy and its closeness to that of the LQR controller depends on the accuracy,of the corresponding a priori information. Since this information is not accurate in practice, a method is presented to evaluate the deviation of the performance of the decentralized system from that of its centralized counterpart. Furthermore, the necessary and sufficient conditions for the stability of the overall closed-loop system in presence of parameter perturbations are given through a series of simple tests. It is shown that stability of the overall system is independent of the magnitude of the expected value of the initial states. One of the advantages of the proposed decentralized strategy over its centralized counterpart is that it is in general more robust to the uncertainties in the system model. Simulation results demonstrate the effectiveness of the proposed controller in terms of feasibility and performance.