| 摘要: |
| 为解决多无人机路径规划鲁棒性不足、收敛速度慢的问题，提出了一种分布式多无人机路径规划方法。首先，将机队成员视为具有本地控制器的非线性动态系统，并将多无人机控制视为在线优化问题构建路径优化成本函数模型。其次，通过将局部控制和分布式路径规划相结合，并使用粒子群优化算法将平移运动与旋转动力学解耦，以实现对多无人机的分布式路径规划和有效控制。最后，通过建立基于四元数的无人机动力学模型与其他两种方法进行对比仿真测试，并搭建由三架无人机组成的机队进行飞行测试，验证所提方法的有效性。实验结果表明，所提方法能够实现对多无人机的最佳路径规划和有效控制，并具有良好的抗干扰性能，且与其他对比方法相比具有更好的收敛性、鲁棒性和抗振荡性。 |
| 关键词: 多无人机 路径规划 粒子群优化(PSO) 分布式控制 |
| DOI：10.20079/j.issn.1001-893x.220427008 |
|
| 基金项目:国网电网有限公司科技项目(5600-202019167A-0-0-00) |
|
| A Multi-UAV Distributed Path Planning Algorithm |
| LIU Kaifen,FENG Ye,ZHANG Feixia,CHEN Ye |
| (1.School of Computer Science,Southwest Jiaotong University,Chengdu 611756,China;2.School of Artificial Intelligence,Chongqing Science and Technology Vocational College,Chongqing 402160,China;3.Intelligent Control Division,Zhejiang Huayun Technology Co.,Ltd.,Hangzhou 310056,China;4.Marketing Service Center,State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050081,China) |
| Abstract: |
| In order to solve the problems of insufficient robustness and slow convergence speed of Unmanned Aerial Vehicle（UAV） fleet path planning,and to improve the effectiveness and fault tolerance of UAV fleet path planning,a distributed UAV fleet path planning method is proposed.First,the fleet members are regarded as nonlinear dynamic systems with local controllers,and the UAV fleet control is regarded as an online optimization problem to construct a path optimization cost function model.Second,by combining local control and distributed path planning,and using Particle Swarm Optimization(PSO) algorithm to decouple translational motion from rotational dynamics,distributed path planning and effective control of UAV fleets are achieved.Finally,the effectiveness of the proposed method is verified by establishing a quaternion-based UAV dynamics model and the other two methods for comparative simulation tests,and a fleet of three UAVs for flight tests is built.The experimental results show that the proposed method can achieve the optimal path planning and effective control of the UAV fleet,and has good anti-jamming performance.Compared with other comparative methods,the proposed method has better convergence,robustness and anti-oscillation. |
| Key words: UAV fleet path planning particle swarm optimization(PSO) distributed control |
