Distributed Optimal Defensive Trajectory Planning against Adversarial Swarms with Receding Horizon

This paper presents a distributed optimal trajectory planning algorithm for defending drones against adversarial entities that attack a high-value unit (HVU). A single defending drone by itself cannot achieve the collective objective, which is to maintain the survival probability of the HVU close to one. This is because the objective depends on the other defending drones' strategies as well. The proposed Nash equilibrium seeking distributed optimization enables each defending drone to calculate its own optimal trajectory exchanging information with its neighbors, which leads to achieving the collective objective. Further, since the attacking drones' dynamics are not exactly known in general, each defending drone calculates and updates its optimal trajectory at every sampling time with a receding time horizon.