基于柱状空间和支持向量机的无人机巡线避障方法  

Obstacle Avoidance For UAV Power Line Inspection Based on Cylindrical Space and Support Vector Machine

作　　者：张剑 王世勇[2] 陈玺 徐华东[2] 俞登科 杨忠[2] 

机构地区：[1]杭州市萧山区供电公司,浙江杭州311200 [2]南京航空航天大学自动化学院,江苏南京210016

出　　处：《中国电力》2015年第3期56-60,66,共6页Electric Power

基　　金：浙江省2014年群众性科技创新资助项目(5211H1140054);博士点基金(博导类)资助项目(20113218110013);江苏省产学研联合创新资金资助项目(BY2012018)~~

摘　　要：针对电力无人机巡线的避障问题,提出了基于柱状空间法和支持向量机的电力无人机巡线避障方法。首先,根据飞行区域内的障碍物,建立了柱状空间模型;引入支持向量机,从一个全新的角度实现无人机对环境的避障;最后将方法应用于无人机中,并进行了相关的仿真验证。仿真结果表明,所提出的方法可以确保电力无人机巡线时安全可靠地避障。To deal with the obstacle avoidance problems in power line inspection with Unmanned Aerial Vehicle （UAV）, a novel obstacle avoidance approach is proposed for UAV llne inspection based on cylindrical space and support vector machine algorithm. Firstly, the cylindrical space model is established according to the obstacles in the flight zone. Secondly, the support vector machine is introduced to achieve the obstacle avoidance with UAV in a new perspective way. Finally, the novel method is applied in the UAV and tested by simulation. The simulation results illustrate that the proposed method can ensure obstacle avoidance in UAV power line inspection with safety and reliability.

关 键 词：无人机 巡线 障碍物 避障方法 柱状空间模型 支持向量机 飞行环境 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TM75[电气工程—电力系统及自动化]