Secure LQG control for a Quadrotor under false data injection attacks

The linear quadratic Gaussian (LQG) control for a quadrotor unmanned aerial vehicle (UAV) under false data injection attacks is studied. The LQG control depends on optimal state estimation, while this type of attacks makes the optimal state estimation unimplementable in practice. To address this problem, the authors prose a framework to detect attacks and then augment the information for controller design. Based on this framework, the LQG control is designed and a sufficient condition for the security of the closed-loop control system is given. Finally, the LQG controller is applied to a quadrotor UAV, and some experiments are carried out to illustrate its effectiveness.