首页|基于深度强化学习的无人机集群实时航路规划

基于深度强化学习的无人机集群实时航路规划

扫码查看
针对无人机城市任务复杂化场景,提出了一种基于深度强化学习的无人机集群实时航路规划算法,通过对威胁、障碍物等进行统一化建模,构建集群航路规划模型,将无人机传感器探测信息作为深度神经网络的状态输入,优化了无人机航路规划的能耗和成功率,通过仿真,证明了算法的有效性和实用性.
Real-time route planning for UAV swarm based on DDQN
A DDQN-based real-time route planning algorithm is proposed for UAV swarm navigation for ur-ban mission complex environments.Firstly,the UAV swarm route planning system model is given by unified modeling of threats,obstacles,etc.Next,the UAV sensor detection information is used as the state input of the deep neural network.Finally,the energy cost and success rate of UAV route planning are optimized.Simulation results show that the algorithm is effective and practical.

UAV swarmroute planningDDQNurban mission

宋海伟、栗志、田达、吕丹阳、吴克钊、黄金磊

展开 >

南京电子设备研究所,江苏 南京 210007

无人机集群 航路规划 深度强化学习 城市任务

2024

航天电子对抗
中国航天科工集团公司8511研究所

航天电子对抗

影响因子:0.382
ISSN:1673-2421
年,卷(期):2024.40(5)