| 多智能体强化学习飞行路径规划算法 |
| |
| 引用本文： | 李东华,江驹,姜长生.多智能体强化学习飞行路径规划算法[J].电光与控制,2009,16(10):10-14. |
| |
| 作者姓名： | 李东华 江驹 姜长生 |
| |
| 作者单位： | 南京航空航天大学自动化学院,南京,210016 |
| |
| 摘 要： | 为了减轻现代空战中大量信息处理给飞行员带来的负担,同时为了实现无人机航路自主规划,提出了一种基于多智能体强化学习理论的飞行路径规划算法.该算法采用多智能体强化学习的方法,采用两个功能不同的智能体,分别对应局部和全局路径规划.该算法对状态和动作空间进行划分和抽象,有效地减少了状态的数量,解决了强化学习维数灾难的问题.最后用Matlab对此算法进行了数字仿真,验证了算法的可行性,仿真实验结果显示该算法收敛速度快,能够解决飞行路径规划的任务.
|
| 关 键 词： | 多智能体系统 强化学习 路径规划 无人机 自主规划 |
| 收稿时间： | 2008/10/9 |
A Flight Path Planning Algorithm Based on Multi-Agent Reinforcement Learning Method |
| |
| LI Donghua,JIANG Ju,JIANG Changsheng.A Flight Path Planning Algorithm Based on Multi-Agent Reinforcement Learning Method[J].Electronics Optics & Control,2009,16(10):10-14. |
| |
| Authors: | LI Donghua JIANG Ju JIANG Changsheng |
| |
| Affiliation: | LI Donghua,JIANG Ju,JIANG Changsheng(College of Automation Engineering,Nanjing University of Aeronautics , Astronautics,Nanjing 210016,China) |
| |
| Abstract: | To reduce the working load of pilots caused by processing the large amount of information in air combat,and also to implement path planning for Unmanned Aerial Vehicles(UAV),a flight path planning algorithm based on multi-agent reinforcement learning method is proposed in this paper.The multi-agent reinforcement learning method is adopted in this algorithm,and two agents with different function are used to deal with the local and global path planning respectively.The state and action spaces are divided and ... |
| |
| Keywords: | multi-agent system reinforcement learning path planning UAV autonomous planning |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
