基于路径像素分布统计导引的无人机自主循迹  

UAVs’Autonomous Tracking Flight Based on Pixel Distribution Statistical Guidance

作　　者：冯宇 崔峰 高东 Feng Yu;Cui Feng;Gao Dong(National Space Secience Center,Beijing 100000;University of Chinese Academy of Sciences,Beijing 100000)

机构地区：[1]国家空间科学中心,北京100000 [2]中国科学院大学,北京100000

出　　处：《现代计算机》2021年第29期28-35,共8页Modern Computer

基　　金：北京市科技计划_空间科学国家实验室培育项目(Z191100004319004)。

摘　　要：文章研究无人机利用视觉传感器自主循迹飞行,针对如何规划不同形状路段的目标序列点的问题,提出了从实时采集图像中提取路径后分块统计像素来确定局部目标路径点坐标的算法。首先从灰度图像中提取路径区域的信息;然后根据灰度图像栅格化后像素分布统计结果确定局部路径序列点坐标;最后根据规划结果计算控制指令。设计了包含不同拐角的路径来检验算法,飞行测试结果表明算法能达到使无人机沿不同形状路径循迹飞行的目的。This paper focuses on the research of drones using visual sensers to autonomously flight along a designated path.Aim⁃ing at solving the problem of planning the target sequence points of different shapes road sections,this paper proposed the algorithm that adding up pixels by area to determine the local target path point coordinates after extracting the path from the real-time acquisition image.First,the algorithm extracted the information of the path area the image gray;then the algorithm calculated the local path se⁃quence point coordinates according to the grid pixel distribution statistics after rasterized the gray image;and finally according to the path planning result,it calculated control instructions.A path with different shapes of corners was designed to test the algorithm.The flight test results showed that the algorithm can achieve the purpose of UAVs tracking flight along paths of different shapes.

关 键 词：无人机 路径识别 路径规划 循迹飞行 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TP391.41[自动化与计算机技术—计算机应用技术]