基于序列图像的固定翼无人机着陆跑道识别技术与跟踪算法验证  

Verification of Landing Runway Identification Technology and Tracking Algorithm of Fixed-wing UAV Based on Sequence Image

作　　者：张卫星[1] Zhang Weixing(Suzhou Institute of Industrial Technology,Suzhou 215104,China)

机构地区：[1]苏州工业职业技术学院,江苏苏州215104

出　　处：《科学技术创新》2022年第27期179-182,共4页Scientific and Technological Innovation

基　　金：苏州科技发展规划项目“基于人工智能的植物病虫害识别系统的研制”(SNG201917)。

摘　　要：提出了一种基于序列图像的固定翼无人机着陆跑道识别技术,并重点从图像预处理、图像边缘检测、跑道特征直线提取与跑道识别、相对位置估算等方面,概述了无人机动态飞行中识别跑道的基本流程和关键技术;随后用一台SZD-45型固定翼无人机进行了飞行实验,对跑道识别与跟踪算法进行了验证。实验结果表明,实时压缩跟踪算法能够较为准确地获取目标坐标信息与相对航向角,在准确地识别跑道边缘的基础上估算无人机与跑道中线的相对位置,实现了无人机安全着陆。This paper proposes a fixed-wing UAV landing runway recognition technology based on sequence images, and focuses on image preprocessing, image edge detection, runway feature line extraction and runway recognition, and relative position estimation. The basic process and key technology of identifying the runway in the dynamic flight of the aircraft were carried out;then the flight experiment was carried out with a SZD-45 fixed-wing unmanned aerial vehicle, and the runway identification and tracking algorithm was verified.The experimental results show that the real-time compression tracking algorithm can accurately obtain the target coordinate information and relative heading angle, and estimate the relative position of the UAV and the centerline of the runway on the basis of accurately identifying the edge of the runway, thus realizing the safe landing of the UAV.

关 键 词：无人机 跑道识别技术 图像边缘检测 压缩跟踪算法 

分 类 号：V279[航空宇航科学与技术—飞行器设计]