无人机巡检架空输电线路自主避障导航算法研究  

Research on Autonomous Obstacle Avoidance Navigation Algorithm for UAV Patrol Overhead Transmission Line

作　　者：张红兵 索春光[1] 宁友欢 张文斌[2] ZHANG Hong-bing;SUO Chun-guang;NING You-huan;ZHANG Wen-bin(Faculty of Science,Kunming University of Science and Technology,Kunming 650504,China;Faculty of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Kunming 650504,China)

机构地区：[1]昆明理工大学理学院,云南昆明650504 [2]昆明理工大学机电工程学院,云南昆明650504

出　　处：《软件》2020年第10期236-241,共6页Software

摘　　要：随着无人机技术的发展,无人机被广泛应用于高压输电线路进行巡检,但受到环境因素的影响作业人员对无人机的控制难度较大,甚至导致无人机撞向输电导线或者铁塔引发电力事故。因此,提高无人机的自主巡航能力是无人机巡检高压输电线的重要技术,基于此,本文提出一种基于模糊神经网络多传感器数据融合方法,实现无人机对架空输电线路的自主避障导航。该方法首先将需要巡检的输电线路的坐标信息输入无人机飞控系统,实现通过铁塔坐标对无人机进行全局导航,而在局部则采用虚拟势场法实现对无人机自主避障。对于多传感器数据分析,本文将模糊神经网络结合模糊聚类的方法进行改进提出一种多传感器数据融合算法,实现对无人机周围环境的准确采集,从而实现对无人机巡线的自主避障导航。最后构建仿真实验平台,对无人机寻线自主避障导航进行模拟实验,从而验证了该算法的可行性。With the development of UAV technology,UAVs are widely used in high-voltage transmission lines for inspection,However,under the influence of environmental factors,it is more difficult for the operator to control the drone,and even causes the drone to hit the transmission line or the tower to cause an electrical accident.Therefore,improving the autonomous cruising capability of the UAV is of great significance for the UAV to inspect the high-voltage transmission lines.Based on this,this paper proposes a multi-sensor data fusion method based on fuzzy neural network to achieve autonomous obstacle avoidance navigation of UAVs on overhead transmission lines.This method first inputs the coordinate information of the transmission line that needs to be inspected into the UAV flight control system to achieve global navigation of the UAV through the tower coordinates,and locally uses the virtual potential field method to achieve autonomous obstacle avoidance for the UAV.For the analysis of multi-sensor data,this paper improves the method of fuzzy neural network combined with fuzzy clustering and proposes a multi-sensor data fusion algorithm to achieve accurate collection of the surrounding environment of the drone,thereby achieving autonomous avoidance of the UAV line patrol Barrier navigation.Finally,a simulation experiment platform was constructed to carry out simulation experiments on the autonomous liner avoidance navigation of the UAV,thus verifying the feasibility of the algorithm.

关 键 词：无人机寻线 多传感器数据融合 自主避障导航 模糊神经网络 自适应 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]