\r侯永宏¹，刘 艳¹，吕华龙¹，吴 琦¹，赵 健²，陈艳芳\r²\r
\r
AuthorsHTML:\r侯永宏¹，刘 艳¹，吕华龙¹，吴 琦¹，赵 健²，陈艳芳\r²\r
\r
AuthorsListE:\rHou Yonghong¹，Liu Yan¹，Lü Hualong¹，Wu Qi¹，Zhao Jian²，Chen Yanfang\r²\r
\r
AuthorsHTMLE:\rHou Yonghong¹，Liu Yan¹，Lü Hualong¹，Wu Qi¹，Zhao Jian²，Chen Yanfang\r²\r
\r
Unit:\r\r1. 天津大学电气自动化与信息工程学院，天津 300072；\r
\r\r2. 天津航天中为数据系统科技有限公司，天津 300072\r
\r
\r
Unit_EngLish:\r1. School of Electrical and Information Engineering，Tianjin University，Tianjin300072，China；
2. Tianjin Zhongwei Aerospace Data System Technology Co.，Ltd.，Tianjin 300072，China\r
\r
Abstract_Chinese:\r针对无人机实时自主导航问题，本文设计并实现了一种能自主感知未知室外环境，实时自动规划路径的旋翼无人机系统．首先利用双目视觉，使用经光束法平差(BA)优化的经典SLAM 系统，ORB SLAM2 算法获取无人机位姿信息；再以“推扫式”感知方法和改进的绝对误差和(SAD)算法获取环境信息和障碍物点．其次，结合无人机位姿信息与环境障碍物点生成局部障碍物地图，同时使用并行计算框架，提高系统性能．针对无人机系统实时性问题，设计的SAD 算法只关注固定视差大小的像素块的稀疏匹配．最后，根据生成的当前局部环境障碍物地图与本地轨迹库，自主选择运动轨迹，有效自主规避障碍物，达到实时局部路径规划的效果．以上功能全部在无人机搭载的嵌入式处理器Nvidia Jetson TX2 中完成处理．仿真与实际飞行实验表明：设计的系统基本实现无人机在未知室外场景下的实时自主感知与路径规划，在采集视频分辨率为1 280×720 时，处理速度能达到60 帧/s，为完善低成本无人机的避障与导航功能提供一种参考．\r
\r
Abstract_English:\rUnmanned aerial vehicles(UAVs) have wide application in wilderness search and rescue，environmental exploration，and other fields. UAVs with autonomous navigation functions have been recently reported. To mitigate the real-time autonomous navigation problem of UAVs，a rotor UAV system was designed and implemented herein. The designed UAV system could autonomously perceive the unknown outdoor environment and realize real-time trajectory planning. The bundle adjustment optimized(BA-optimized) ORB SLAM2 algorithm was first used to obtain the posture information of UAV using a binocular camera. The“push-broom”perception method and modified sum of absolute differences(SAD) algorithm were then used to obtain environmental information and obstacle points. Local obstacle maps were constructed based on the environmental obstacle points and UAV posture information using a parallel calculation framework(CUDA) to improve the system performance. To solve the real-time autonomous navigation problem of UAVs，the modified SAD algorithm only focuses on the sparse matching of pixel blocks with a fixed parallax size. According to the generated current local obstacle map and local trajectory library，the motion track can be independently selected to achieve the effect of the real-time local trajectory planning. All the above functions were processed in the embedded NVIDIA Jetson TX2 processor equipped with a drone. Simulation and actual flying experiments show that the designed system realizes real-time autonomous perception and trajectory planning function of UAVs in unknown outdoor scenes. When the resolution of the captured video is 1 280×720，the processing speed can reach 60 frames per second. In summary，this design provides a reference for improving the obstacle avoidance and navigation function of low-cost UAVs.\r
\r
Keyword_Chinese:双目视觉；视觉里程计；并行计算；局部路径规划；自主导航\r

Keywords_English:binocular vision；visual odometer；parallel calculation；local trajectory planning；autonomous navigation\r


PDF全文下载地址:http://xbzrb.tju.edu.cn/#/digest?ArticleID=6375