通信限制下多无人机协同全覆盖路径规划  

Multi UAV Collaborative Full Coverage Path Planning under Communication Limitations

作　　者：姚文军 邓存宝 范楠 张鹏莉 YAO Wen-jun;DENG Cun-bao;FAN Nan;ZHANG Peng-li(College of Safety&Emergency Management Engineering,Taiyuan University of Technology,Jinzhong 030600,China)

机构地区：[1]太原理工大学安全与应急管理工程学院,山西晋中030600

出　　处：《计算机仿真》2024年第12期60-66,共7页Computer Simulation

摘　　要：为解决多无人机区域协同覆盖的问题,提出一种基于奖励函数的实时协同覆盖算法。首先建立环境地图模型,根据先验信息划分为:待覆盖区域、已覆盖区域、边界区域以及禁飞区域;并采用矩阵运算实现地图的快速更新。根据无人机的动力学特性,建立无人机的运动模型。根据动态窗口预测无人机的路径,结合地图信息使之实现障碍物回避,并在陷入困境时,利用图像处理技术和改进A*算法引导无人机飞行。最后进行了仿真,结果表明提出的协同覆盖算法有很高的鲁棒性和覆盖效率。当运行到1200步长时,所提算法的覆盖率为:96.66%,相较于贪婪算法、预测算法和无引导算法分别提高了:33.7%、30.5%和16.05%。In order to resolve the problem of multi-UAV area cooperative coverage,a cooperative coverage algorithm based on real-time rewards is proposed.First,environmental map is established to describe diverse area:uncovered area,covered area,boundary area and no-fly zones.the rapid update of map is realized by using matrix operation.Second,building UAV models to satisfy the maximum turn radius restriction.Dynamic Window Method is employed to predicting the future path.Combining map information to achieve obstacle avoidance.Using image processing techniques and improved A∗algorithms to guide UAV when in trouble.The simulation results demonstrate the efficiency and robustness of the proposed cooperative approach.When running to 1200 steps,the coverage rate of this paper algorithm is:96.66%,compared with greedy algorithm,prediction algorithm and unguided algorithms respectively increased by:33.7%,30.5%and 16.05%.

关 键 词：无人机 路径规划 协同覆盖 图像处理 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TP18[自动化与计算机技术—控制理论与控制工程]