基于深度学习的低空小型无人机障碍物视觉感知  

Obstacle visual sensing based on deep learning for low-altitude small unmanned aerial vehicles

作　　者：寇展 吴健发 王宏伦[1,2] 阮文阳 Zhan KOU;Jianfa WU;Honglun WANG;Wenyang RUAN(School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China;Science and Technology on Aircraft Control Laboratory,Beihang University,Beijing 100191,China;Shenyuan Honors College of Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学自动化科学与电气工程学院,北京100191 [2]北京航空航天大学飞行器控制一体化技术重点实验室,北京100191 [3]北京航空航天大学高等理工学院,北京100191

出　　处：《中国科学：信息科学》2020年第5期692-703,共12页Scientia Sinica(Informationis)

基　　金：国家自然科学基金(批准号:61175084,61673042)资助项目。

摘　　要：本文提出了一种基于深度学习与目标跟踪方法,综合单目视觉和双目立体视觉特点的无人机障碍物实时感知方法.首先,采用深度学习的方法对相机采集的首帧图片进行障碍物检测与识别,并采用目标跟踪算法对首帧检测结果进行实时跟踪,以提高检测系统实时性;同时,使用双目立体视觉技术对当前帧图像进行全图的三维重构,得到环境的空间信息;之后,通过在检测结果的感兴趣区域内进行点云聚类提取等策略并结合上述感知到的信息进行信息融合,得到障碍物的种类、空间位置及轮廓大小.最后,开发实物样机对方法进行验证,结果表明通过采用该方法,无人机在搭载一个双目相机的情况下即可完成对障碍物的实时感知.An obstacle real-time sensing method,which is based on deep learning and target tracking method and integrated with monocular vision and binocular vision,is proposed for unmanned aerial vehicles(UAVs)in this paper.Firstly,it uses the deep learning method to detect and recognize the first-frame figure collected by cameras.Then,it uses the target tracking algorithm to track the detection results for the first-frame figure in real time to improve the real-time performance of the detection system.Meanwhile,it uses the binocular vision technology to execute the three-dimensional reconstruction for the current frame of the entire figure to obtain the environmental spatial information.Subsequently,combined with the points clustering strategy and the information fusion method,it can resolve the types,spatial locations,and outlines of obstacles.Finally,to verify the proposed method,we developed a physical prototype,and the results showed that the real-time sensing for obstacles can be realized under the condition that UAVs are equipped with one binocular camera.

关 键 词：深度学习 目标跟踪 无人机 障碍感知 双目视觉 

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] TP18[自动化与计算机技术—计算机科学与技术]