Predictive Path Planning Strategy for Multi-UAV Systems against Covert Attacks

This study introduces a predictive path planning framework for controlling multiple Unmanned Aerial Vehicles (UAVs) to follow their formation reference trajectories with minimized tracking and formation errors against covert attacks. Utilizing the leader-follower formation control strategy and the model predictive control (MPC) approach, it proposes predictive attack and defense strategies in the single neighbour and the double neighbours scenarios for a UAV swarm outfitted with the Ultra-Wideband (UWB) and GPS sensors. The covert attacker manipulates GPS data, which is undetectable by the UWB sensor, to maximize the formation error and the deviation between the attacked path and reference path, while the defender aims to reduce the bad effects on the UAV flight. The paper presents two algorithms for real-time optimization of attack and defense strategies and employs the Euler method for discretization.