Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2019, Vol. 53 Issue (5): 880-888    DOI: 10.3785/j.issn.1008-973X.2019.05.008
计算机与控制工程     
集中式光伏电站巡检无人机视觉定位与导航
席志鹏(),楼卓,李晓霞,孙艳,杨强*(),颜文俊
浙江大学 电气工程学院，浙江 杭州 310027
Vision-based localization and navigation for UAV inspection in photovoltaic farms
Zhi-peng XI(),Zhuo LOU,Xiao-xia LI,Yan SUN,Qiang YANG*(),Wen-jun YAN
College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China
 全文: PDF(1629 KB)   HTML
摘要：

为了使无人机在集中式光伏电站实现自主飞行，完成光伏组件红外及可见光图像采集任务，针对集中式光伏厂区光伏组串分布特点，提出光伏组串边缘检测方法，并通过视线导引法实现无人机路径跟随控制. 同一厂区的不同光伏组件之间存在颜色差异，针对其颜色特征提出自定义分割方法，结合形状特征可有效识别光伏组件；提取光伏组串轮廓和边缘信息可获取无人机理论飞行路径，通过视线导引法实现无人机对理论飞行路径的精准跟随以确保图像数据采集的有效性和完整性. 实验结果表明，提出的光伏组串识别算法具有较好的适应性和实时性，能够用于无人机理论飞行方向与无人机和光伏组串间偏移量的计算，利用导航控制算法能够实现理想的光伏组串循迹. 光伏组串识别算法和视线导引法能分别有效实现定位和导航，2个程序的结合能够满足无人机飞行控制要求.
关键词： 光伏巡检;  四旋翼无人机(UAV);  视觉定位;  路径跟随;  自主飞行    
Abstract:

In order to achieve autonomous flight for unmanned aerial vehicles (UAVs) in PV farms and complete infrared and visible-light image acquisition, an edge detecting method for photovoltaic (PV) strings was proposed and the line of sight guidance based path following control algorithm was carried out according to the distribution characteristics of PV strings. Color variance widely existes among different PV modules in the same PV farm, based on which a custom color segmentation technique was put forward. This technique could be combined with shape features to realize accurate identification for PV modules. Theoretical flight paths could be gained through contour and edge information of PV strings after which the line of sight guidance method was applied to accurate trajectory tracking control for theoretical path to guarantee the effectiveness and integrity of image data acquisition. Results showed that the proposed recognition algorithm for PV strings was excellent in adaptability and instantaneity and could be used to calculate the theoretical flight direction of UAV and the offsets between the UAV and PV strings, and the ideal trajectory tracking for PV strings can be realized through the navigation control algorithm. The identification approach for PV strings and the line of sight guidance method can work well in localization and navigation respectively, and combination of the two procedures can meet the requirements of UAV flight control.
Key words: photovoltaic inspection    quad-rotor unmanned aerial vehicle (UAV)    vision localization    path following    autonomous flight
收稿日期: 2018-06-26 出版日期: 2019-05-17
CLC:  TP 29  
通讯作者: 杨强     E-mail: xizhipeng@zju.edu.cn;qyang@zju.edu.cn
作者简介: 席志鹏（1993—），男，硕士，从事无人机光伏巡检研究. orcid.org/0000-0002-2394-4288. E-mail： xizhipeng@zju.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
席志鹏
楼卓
李晓霞
孙艳
杨强
颜文俊

引用本文:

席志鹏,楼卓,李晓霞,孙艳,杨强,颜文俊. 集中式光伏电站巡检无人机视觉定位与导航[J]. 浙江大学学报(工学版), 2019, 53(5): 880-888.

Zhi-peng XI,Zhuo LOU,Xiao-xia LI,Yan SUN,Qiang YANG,Wen-jun YAN. Vision-based localization and navigation for UAV inspection in photovoltaic farms. Journal of ZheJiang University (Engineering Science), 2019, 53(5): 880-888.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2019.05.008    或    http://www.zjujournals.com/eng/CN/Y2019/V53/I5/880

图 1  集中式光伏电站区域划分示意图
图 2  无人机导航方法流程图
图 3  光伏组件典型颜色特征
图 4  4种光伏组件RGB颜色空间直方图分布
图 5  自定义规则下光伏组件颜色特征直方图
图 6  光伏组串识别过程
图 7  光伏组串轮廓
图 8  光伏组串边缘直线检测结果
图 9  导航坐标系下的角度关系
图 10  无人机路径跟随过程示意图
图 11  无人机飞行速度示意图
图 12  无人机转弯过程示意图
图 13  不同场景下光伏组件识别结果以及光伏组串边缘检测效果
特征时间 T/ms
颜色特征提取时间 20
形状特征提取时间 15
组串定位时间 4
边缘直线检测时间 12
总时间 51
表 1  图像处理算法实时性分析
图 14  基于视线导引法的无人机导航控制过程仿真结果
图 15  光伏组件分布与无人机飞行路径俯视图
图 16  无人机与光伏组件偏移量变化趋势
1 国家能源局. 国家能源局新闻发布会介绍2017年度相关能源情况等[EB/OL]. (2018-01-24)[2018-06-20]. http://www.nea.gov.cn/2018-01/24/c_136921015.htm.
2 YANG T W, YIN H, RUAN Q Q, et al. Overhead power line detection from UAV video images [C]// Mechatronics and Machine Vision in Practice. Auckland: IEEE, 2013: 74-79.
3 SARAPURA J A, ROBERTI F, CARELLI R, et al. Passivity based visual servoing of a UAV for tracking crop lines [C]// XVII Workshop on Information Processing and Control. Mar del Plata: IEEE, 2017: 1-6.
4 BRAND?O A S, MARTINS F N, SONEGUETTI H B. A vision-based line following strategy for an autonomous UAV [C]// International Conference on Informatics in Control, Automation and Robotics. Colmar: IEEE, 2015: 314-319.
5 HUANG H, ZHOU H, QIN H D, et al. Underwater vehicle visual servo and target grasp control [C]// IEEE International Conference on Robotics and Biomimetics. Qingdao: IEEE, 2017: 1619-1624.
6 HWANG C L, HUANG H H. Experimental validation of a car-like automated guided vehicle with trajectory tracking, obstacle avoidance, and target approach [C]// Conference of the IEEE Industrial Electronics Society. Beijing: IEEE, 2017: 2858-2863.
7 MARTINEZ C, SAMPEDRO C, CHAUHAN A, et al. Towards autonomous detection and tracking of electric towers for aerial power line inspection [C]// International Conference on Unmanned Aircraft Systems. Orlando: IEEE, 2014: 284-295.
8 LIU Y S, WANG Q X, HU H S, et al A novel real-time moving target tracking and path planning system for a quadrotor UAV in unknown unstructured outdoor scenes[J]. IEEE Transactions on Systems Man and Cybernetics Systems, 2018, 1- 11
9 LEIRA F S, JOHANSEN T A, FOSSEN T I. A UAV ice tracking framework for autonomous sea ice management [C]// International Conference on Unmanned Aircraft Systems. Miami: IEEE, 2017: 581-590.
10 YAN F L, LI B Q, SHI W X, et al Hybrid visual servo trajectory tracking of wheeled mobile robots[J]. IEEE Access, 2018, 6: 24291- 24298
doi: 10.1109/ACCESS.2018.2829839
11 XIAO S L, LI Y M Visual servo feedback control of a novel large working range micro manipulation system for microassembly[J]. Journal of Microelectromechanical Systems, 2014, 23 (1): 181- 190
doi: 10.1109/JMEMS.2013.2268849
12 ZHANG D, WU F J, LI X X, et al. Aerial image analysis based on improved adaptive clustering for photovoltaic module inspection [C]// International Smart Cities Conference. Wuxi: IEEE, 2017: 1-6.
13 WU F J, ZHANG D, LI X X, et al. Aerial image recognition and matching for inspection of large-scale photovoltaic farms [C]// Smart Cities Conference. Wuxi: IEEE, 2017: 1-6.
14 初广丽, 张伟, 王延杰, 等 基于机器视觉的水果采摘机器人目标识别方法[J]. 中国农机化学报, 2018, (2): 83- 88
CHU Guang-li, ZHANG Wei, WANG Yan-jie, et al A method of fruit picking robot target identification based on machine vision[J]. Journal of Chinese Agricultural Mechanization, 2018, (2): 83- 88
15 LI X H, ZHAO Y, ZHANG J, et al. A hybrid pso algorithm based flight path optimization for multiple agricultural UAVs [C]// International Conference on TOOLS with Artificial Intelligence. San Jose: IEEE, 2017: 691-697.
16 高海涛. UUV循迹跟踪与环境最优控位方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2014.
GAO Hai-Tao. Research on UUV path following and weather optimal dynamic position [D]. Harbin: Harbin Engineering University, 2014.
[1] 王雅静,王群,李博闻,刘志文,朴媛媛,遇涛. 基于脑电信号预发作数据段选取的癫痫发作预测[J]. 浙江大学学报(工学版), 2020, 54(11): 2258-2265.
[2] 荆丹翔,韩军,徐志伟,陈鹰. 基于成像声呐的水下多目标跟踪研究[J]. 浙江大学学报(工学版), 2019, 53(4): 753-760.
[3] 任逸飞, 陆志强, 刘欣仪, 张猛. 考虑技能水平的多技能资源约束项目调度[J]. 浙江大学学报(工学版), 2017, 51(5): 1000-1006.
[4] 周炳海, 赵猛. 柔性Job Shops集成调度启发式算法[J]. 浙江大学学报(工学版), 2016, 50(6): 1073-1079.
[5] 周炳海, 陈锦祥, 赵猛. 基于晶圆优先级的连续型Interbay搬运系统性能分析[J]. 浙江大学学报(工学版), 2015, 49(2): 296-302.
[6] 赵立杰, 柴天佑, 袁德成, 刁晓坤. 增强操作工况识别可靠性的概率PLS-ELM方法[J]. J4, 2013, 47(10): 1747-1752.
[7] 孙跃, 赵志斌, 苏玉刚, 唐春森. 非接触电能传输系统参数非线性规划[J]. J4, 2013, 47(2): 353-360.
[8] 孟庆龙, 王元. 基于CFD的空间场温度系统建模与控制[J]. J4, 2012, 46(8): 1478-1484.
[9] 赵立杰, 汤健, 柴天佑. 基于选择性极限学习机集成的磨机负荷软测量[J]. J4, 2011, 45(12): 2088-2092.
[10] 夏明, 董亚波, 鲁东明. 无线传感网逐跳自适应FEC传输可靠性
保证方法[J]. J4, 2011, 45(2): 273-279.
[11] 孙宁, 张化光, 王智良. 不确定分数阶混沌系统的滑模投影同步[J]. J4, 2010, 44(7): 1288-1291.
[12] 王建中, 金波, 鲁仁全. 基于渗漏模型的城市排水系统建模与分析[J]. J4, 2010, 44(7): 1382-1386.
[13] 汤健, 赵立杰, 岳恒, 柴天佑. 基于多源数据特征融合的球磨机负荷软测量[J]. J4, 2010, 44(7): 1406-1413.