Formation Control for Quadrotor Swarms with Deception Attacks Via Singular Perturbation
Ying Zhang, Lei Ma, Chunyu Yang, Linna Zhou · 2022 IEEE International Conference on Robotics and Biomimetics (ROBIO) · 2022
This paper is concerned with the formation control problem of quadrotor swarms subject to deception attacks. Malicious attackers disrupt formation missions by injecting false data into the information exchanged between the quadrotors. First, in order to solve the ill-conditioned numerical problem while reducing the computational burden, the quadrotor swarm system is decomposed into two reduced-order slow and fast subsystems by using the singular perturbation decomposition theory. Then, a novel composite control architecture is proposed, which consists of a distributed attack-resilient formation controller for the position slow subsystem, a PID controller for the attitude slow subsystem and a linear quadratic regulation controller for the fast subsystem. It is worth emphasizing that the interaction information between the quadrotors is only processed in the slow subsystem, which greatly improves the calculation speed. Finally, a simulation example with four quadrotors is given to verify the effectiveness of the proposed algorithm.