Task Assignment for UAV Swarm Saturation Attack: A Deep Reinforcement Learning Approach

• DOI: 10.3390/electronics12061292
• Publication Date: 2023-03-09T06:35:37Z
• Authors: Feng Qian; Kai Su; Xin Liang; Kan Zhang

Task assignment is a challenging problem in multiple unmanned aerial vehicle (UAV) missions. In this paper, we focus on the task for UAV swarm saturation attack, which deep reinforcement learning (DRL) framework developed. Specifically, first construct mathematical model to formulate attack and consider it as Markov Decision Process (MDP). We then design policy neural network using attention mechanism. also propose training algorithm based gradient method so that our agent can learn an effective policy. The experimental results have shown DRL generate high-quality solutions different scales, meets requirements of real-time flexibility actual situation.