强化学习
马尔可夫决策过程
计算机科学
群体行为
任务(项目管理)
人工智能
灵活性(工程)
分配问题
粒子群优化
人工神经网络
机器学习
马尔可夫过程
数学优化
工程类
数学
统计
系统工程
作者
Feng Qian,Kai Su,Liang Xin,Kan Zhang
出处
期刊:Electronics
日期:2023-03-08
卷期号:12 (6): 1292-1292
被引量:1
标识
DOI:10.3390/electronics12061292
摘要
Task assignment is a challenging problem in multiple unmanned aerial vehicle (UAV) missions. In this paper, we focus on the task assignment problem for a UAV swarm saturation attack, in which a deep reinforcement learning (DRL) framework is developed. Specifically, we first construct a mathematical model to formulate the task assignment problem for a UAV swarm saturation attack and consider it as a Markov Decision Process (MDP). We then design a policy neural network using the attention mechanism. We also propose a training algorithm based on the policy gradient method so that our agent can learn an effective task assignment policy. The experimental results have shown that our DRL method can generate high-quality solutions for different problem scales, which meets the requirements of real-time and flexibility in the actual situation.
科研通智能强力驱动
Strongly Powered by AbleSci AI