摘 要:面对煤矿中复杂的地形,巡检无人机很难将降落地标与环境区分开,无法实现定点的精准降落。基于以上问题,提出了一种基于视觉导航的煤矿输电线无人机自主降落系统。为了区分地标和环境区,将地标设计为一种拓扑模式,为了提高图像二值化的检测效率,提出了一种动态阈值方法。根据有效图像信息计算水平面内的相对距离,用线性插值法求出相对高度。在相应平台上进行降落实验,实验结果表明,提出的自主降落系统性能稳定,可以准确自主降落。
关键词:计算机视觉;图像二值化;无人机;自主降落
中图分类号:V279;V249;TP391.41 文献标识码:A 文章编号:2096-4706(2020)04-0014-03
Vision Based Autonomous Landing System of UAV for Coal Mine Transmission Line Inspection
ZHANG Chuanjiang1,LIU Xianfeng2,CHEN Banggan2,CAO Jinfang1,MENG Xiangyu3
(1.Huaibei Mining (Group) Co.,Ltd.,Huaibei 235000,China;2.Huaibei Mining Culture and Tourism Media Co.,Ltd.,Huaibei 235000,China;3.Xutuan Mine of Huaibei Mining Co.,Ltd.,Huaibei 235000,China)
Abstract:Facing the complex terrain in the coal mine,it is difficult for UAV to distinguish the landing landmark from the environment and realize the precise landing at a fixed point. Based on the above problems,an autonomous landing system of UAV for coal mine transmission line based on visual navigation is proposed. In order to distinguish the landmark from the environment area,the landmark is designed as a topological mode. In order to improve the detection efficiency of two value image,a dynamic threshold method is proposed. According to the effective image information,the relative distance in the horizontal plane is calculated,and the relative height is calculated by linear interpolation. The landing experiment is carried out on the corresponding platform. The experimental results show that the proposed autonomous landing system has stable performance and can land accurately.
Keywords:computer vision;image binarization;UAV;autonomous landing
参考文献:
[1] 杨婷 . 电力巡检四旋翼无人机自主着陆系统研究 [D]. 成 都:电子科技大学,2019.
[2] FAN R,JIAO J,PAN J,et al. Real-Time Dense Stereo Embedded in A UAV for Road Inspection [C]//IEEE Conference on Computer Vision and Pattern Recognition Workshops,2019.
[3] 索文凯,胡文刚,张炎,等 . 无人机自主降落过程视觉定位方法研究 [J]. 激光技术,2019,43(5):101-106.
[4] FAN R.Real-time computer stereo vision for automotive applications [D].Bristol:University of Bristol,2018.
[5] 徐焕太 . 基于双目视觉的多旋翼无人机自主降落定位方法研究 [D]. 哈尔滨:哈尔滨理工大学,2018.
[6] FAN R,AI X,DAHNOUN N.Road Surface 3D Reconstruction Based on Dense Subpixel Disparity Map Estimation [J].IEEE Transactions on Image Processing,2018,27(6):3025-3035.
[7] 屈小媚,刘韬,谈文蓉 . 基于多无人机协作的多目标无源定位算法 [J]. 中国科学:信息科学,2019,49(5):570-584.
[8] 贾配洋,彭晓东,周武根 . 四旋翼无人机自主移动降落方法研究 [J]. 计算机科学,2017,44(S2):520-523.
[9] 董素河 . 输电线路智能无人机巡检的研究及应用 [J]. 石化技术,2019,26(8):350-351.
作者简介:张传江(1969-),男,汉族,安徽淮北人,淮北矿业机电装备处处长,高级工程师,本科,研究方向:矿山机电。