LiDAR and visual information fusion position system of living work robot for distribution network

doi:10.13474/j.cnki.11-2246.2021.0052

Abstract

Abstract: This paper proposes a system based on single-line LiDAR sensor and vision fusion method to detect the spatial three-dimensional coordinates of the wire. It can improve the recognition and positioning accuracy of live working robot for distribution network effectively. LiDAR and visual information fusion position system based on radar sensors and visual fusion method to calculate the spatial three-dimensional coordinates of the wire. Living work robot can work outdoor with bright sunlight environment by using the multiple sensor fusion algorithm. At first, the LiDAR and the camera should be calibrated in advance to make the image pixels correspond to the LiDAR depth point cloud. Then, the paper uses Canny algorithm and Hough transform to calculate the position of wire. Finally, the robot can recognise and position wire in high precision and high efficiency under outdoor light.

Key words: LiDAR, computer vision, living work robot for distribution network, calibration, DBSCAN

CLC Number:

P237

REN Qingting, LI Shuai, Lü Peng, ZHANG Tong. LiDAR and visual information fusion position system of living work robot for distribution network[J]. Bulletin of Surveying and Mapping, 2021, 0(2): 98-102,116.

References

[1] 王凯军,应鸿,鲁守银,等.高压带电作业机器人的开发背景及展望[J].浙江电力,2005(6):15-16.
[2] SAWADA J,KUSUMOTO K,MAIKAWA Y,et al. Mobile robot for inspection of power transmission lines[J].IEEE Trans on Power Delibery,1991,6(1):309-315.
[3] NAKASHIMA M,YANO K,MARUYAMA Y,et al.The hot line work robot system "Phase Ⅱ" and its human-robot interface "MOS"[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems.[S.l.]:IEEE, 1995:116-123.
[4] BROGGI A,BERTOZZI M,FASCIOLI A,et al. Visual perception of obstacles and vehicles for platooning[J].IEEE Transactions on Intelligent Transportation System,2000,1(3):164-176.
[5] ESTER M, KRIEGEL H P, SANDER J, et al.A density-based algorithm for discovering clusters in large spatial databases with noise[C]//Proceedings of the 2nd ACM SIGKDD.Portland:ACM, 1996:226-231.
[6] 关超华,陈泳丹,陈慧岩,等.基于改进DBSCAN算法的激光雷达车辆探测方法[J].北京理工大学学报,2010,30(6):732-736.
[7] 麦新晨,杨明,王春香,等.一种基于多传感器融合的车辆检测与跟踪方法[J].上海交通大学学报,2011,45(7):1012-1016.
[8] 周旭.基于改进粒子滤波的算法研究[D].南京:南京理工大学,2014.
[9] 彭梦,蔡自兴.一种基于双平行平面的激光雷达和摄像机标定方法[J].中南大学学报(自然科学版),2012,43(12):4735-4742.
[10] 刘洋, 杨必胜, 梁福逊. 机载激光点云中高压电塔自动识别方法[J]. 测绘通报, 2019(1):34-38.
[11] 张省, 朱伟. 基于SIFT匹配和RANSAC算法的超分辨率重建[J]. 测绘通报, 2019(10):119-122.
[12] 蒋科迪, 殷勇, 范开桂, 等. 基于Canny算子的南通江海岸线研究[J]. 测绘通报, 2019(10):83-88.
[13] 王光,林国宇.改进的自适应参数DBSCAN聚类算法[J].计算机工程与应用,2020,56(14):45-51.
[14] 龚昕,张楠.基于Hough变换的圆检测算法的改进[J].信息技术,2020,44(6);89-93,98.
[15] 张秋雁,杨忠,陈聪,等.一种复杂环境下的电力线检测方法[J].应用科技,2019,46(5):51-56.