“碳达峰、碳中和”背景下，煤炭作为传统化石能源消耗的主体，我国能源安全的压舱石，煤矿智能化建设是大势所趋。目前煤矿综采工作面采用人工巡检，或者借助视频完成巡检，均无法满足综采工作面智能巡检需求。本文以课题组前期设计的煤矿综采工作面柔性轨道式巡检机器人为基础，提出一种基于柔性轨道状态识别的综采工作面巡检机器人控制系统设计方案，借助机器视觉、机器人控制、图像处理以及传感器信息处理等技术，实现综采工作面正常巡检以及柔性轨道式巡检机器人自主运行，为综采工作面机器人化巡检奠定基础。

针对分段柔性轨道随液压支架推移影响工作面巡检机器人的通过性等问题，提出一种基于视觉的分段柔性轨道状态判别方法。针对工作面采集图像存在光照不均匀、光照过低等问题，提出一种基于光照不均匀校正的图像自适应增强算法对采集图像进行增强处理并对该方法进行主客观评价。将增强后的图像经过畸变校正、滤波、透视变换等方法提取轨道图像的ROI；研究Sobel边缘检测算法以及概率霍夫变换实现柔性连接机构边缘检测，利用K-means算法实现边缘左右侧分类，结合RANSAC算法拟合柔性连接机构边缘，获得柔性连接机构的弯折角度，实现分段柔性轨道的状况判别，并完成柔性连接机构状况分类，为实现巡检机器人自主运行奠定基础。

针对工作面巡检机器人通过不同工况柔性连接机构时的运行速率调节和速度稳定性问题，提出一种模糊PID算法与加减速算法相结合的运动控制方案。通过对工作面巡检机器人进行动力学分析与仿真，掌握巡检机器人在柔性轨道上的运动变化规律，并提出了模糊PID算法与加减速算法结合的巡检运动控制策略，分析对比了不同加减速算法的优缺点，决定选用S型加减速算法。通过对比实验验证了控制算法可行性。

巡检机器人系统设计要满足巡检性能和视觉处理要求，论文研发一种基于树莓派和STM32的巡检机器人控制系统，对机器人控制系统进行硬件选型和软件详细设计，实现了上位机监控系统和下位机控制系统的相关功能，达到了柔性轨道式机器人工作面巡检要求。

最后搭建工作面柔性轨道式巡检机器人实验平台，对控制系统功能及性能进行验证，系统实现了环境监测、设备监测、数据交互与显示、远程控制、图像增强和自适应控制等功能。结果表明控制系统达到了设计要求，研究对综采工作面智能巡检建设具有一定的借鉴作用。

Under the background of "carbon peaking and carbon neutrality", as the main body of traditional fossil energy consumption and the ballast of China's energy security, the intelligent development and utilization of coal is the general trend. Existing inspection methods such as manual inspection and inspection with the help of video are unable to meet the demand for intelligent review of the fully mechanized coal mining face. This paper proposed a control system design scheme for inspection robots based on flexible track state recognition, based on the group's previous design of flexible track inspection robot for fully mechanized coal mining faces. With the help of machine vision, robot control, image processing, and sensor information processing, the system realized routine inspection of the work face and the autonomous operation of the flexible track inspection robot, laying the foundation for the robotization of the inspection in fully mechanized coal mining face.

A visual method was proposed to distinguish the state of the segmented flexible track, which affected the passage of the inspection robot at the working face when the segmented flexible track was pushed with the hydraulic support. Because of the characteristics of the acquired images, such as uneven illumination and low illumination, this paper proposed an image adaptive enhancement method based on the correction of uneven illumination to enhance the images and evaluated the way subjectively and objectively. Then, the enhanced image was used to extract the ROI of the track image by distortion correction, filtering, perspective transformation, and other methods. The Sobel edge detection algorithm and Probabilistic Hough Transform were studied to realize the edge detection of the flexible connection mechanism. The K-means algorithm was used to classify the left and right sides of the edge. The RANSAC algorithm was used to fit the edge of the flexible connection mechanism, and the bending angle of the flexible connection mechanism was obtained. The condition recognition of the segmented flexible track was realized, and the condition classification of the flexible connection mechanism was completed, which laid the foundation for the autonomous operation of the inspection robot.

A motion control scheme combining fuzzy PID algorithm with acceleration and deceleration algorithm was proposed to solve the problem of speed regulation and speed stability of the inspection robot through flexible tracks under different working conditions. The motion variation law of the inspection robot on the flexible track was mastered through the dynamic analysis and simulation of the fully mechanized mining face inspection robot. And the inspection motion control strategy combining the fuzzy PID algorithm and the acceleration and deceleration algorithm was proposed. The advantages and disadvantages of different acceleration and deceleration algorithms were analyzed and compared, and the S-type acceleration and deceleration algorithm was selected. Comparative experiments verified the feasibility of the control algorithm.

The design of the inspection robot system should meet the requirements of inspection performance and visual processing. In this paper, a control system of inspection robot based on Raspberry pi and STM32 was developed. The hardware selection and software design of the robot control system were carried out. The related functions of the upper computer monitoring system and the lower computer control system were realized, and the inspection requirements of the flexible track robot in fully mechanized coal mining face were achieved.

Finally, an experimental platform of the flexible track inspection robot for working face was built to verify the function and performance of the control system. The system realizes the functions of environmental detection, equipment monitoring, data interaction and display, remote control, image enhancement, and adaptive control. The results show that the control system meets the design requirements, and the research has a certain reference for the construction of intelligent inspection in fully mechanized coal mining face.

[1]杨洪涛, 李富兵. 中国能源资源开发利用布局与国土资源节约集约的耦合关系分析[J]. 中国矿业, 2016, 25(11): 52-54+98.

[2]王国法, 刘峰. 中国煤炭智能化发展报告(2020)[M]. 北京: 科学出版社, 2020.

[3]谢和平, 吴立新, 郑德志. 2025年中国能源消费及煤炭需求预测[J]. 煤炭学报, 2019, 44(7): 1949-1960.

[4]王国法, 刘峰, 庞义辉, 等. 煤矿智能化: 煤炭工业高质量发展的核心技术支撑[J]. 煤炭学报, 2019, 44(7): 1949-1960.

[5]王国法, 任世华, 庞义辉, 等. 煤炭工业“十三五”发展成效与“双碳”目标实施路径[J]. 煤炭科学技术, 2021, 49(9): 1-8.

[6]潘祥生. 矿用悬线式巡检机器人的设计与动力学分析[J]. 煤矿机械, 2019, 40(11): 4-7.

[7]杜春晖. 综采工作面视频监控系统优化设计[J]. 工矿自动化, 2020, 46(8): 94-100.

[8]黄曾华, 王峰, 张守祥. 智能化采煤系统架构及关键技术研究[J]. 煤炭学报, 2020, 45(06): 1959-1972.

[9]鲁浩. 综合管廊巡检机器人控制系统的设计与实现[D]. 南京: 东南大学, 2018.

[10]葛世荣. 煤矿机器人现状及发展方向[J]. 中国煤炭, 2019, 45(07): 18-27.

[11]Novák P, Babjak J, Kot T, et al. Control system of the mobile robot TELERESCU-ER[C]// Applied Mechanics and Materials, Trans Tech Publications Ltd, 2015, 772: 466-470.

[12]张军辉, 李晓鹏. 基于多传感器的煤矿探测机器人控制系统研究[J]. 煤炭科学技术, 2012, 40(07): 89-92.

[13]王亚男, 李明远. 基于STM32控制的全方位可视煤矿救援机器人的研究设计[J]. 煤炭技术, 2022, 41(01): 242-244.

[14]江新奇, 刘敬玉, 李忠飞, 等. 煤矿巡检机器人智能传感与控制系统设计研究[J]. 煤炭工程, 2022, 54(01): 171-175.

[15]宣鹏程, 孙稚媛, 周东旭, 等. 煤矿轨道式带式输送机巡检机器人系统设计[J]. 煤矿机械, 2020, 41(05): 1-3.

[16]周庆, 肖前波, 万欣, 等. 500Kv输电线路巡检机器人越障控制研究[C]//仪器仪表学报(2015(增刊)第36卷), 2015: 148-152.

[17]毛浩, 薛忠新, 范生军, 等. 张家峁煤矿回风巷道智能巡检机器人系统[J]. 煤矿安全, 2021, 52(07): 107-111.

[18]曹薇, 陈亮. 电气巡检机器人避障导航PLC控制系统设计[J]. 自动化与仪器仪表, 2021(06): 157-160+164.

[19]陈海龙. 煤矿选煤厂巡检机器人的研究与设计[D]. 徐州: 中国矿业大学, 2014.

[20]牛雪娟, 孙宏图. 变倾角自适应双轮自平衡机器人控制算法[J]. 控制工程, 2021, 28(02): 306-312.

[21]唐翠微, 蒲东清. 工业机器人低合金钢焊缝自适应焊接控制方法[J/OL]. 机械设计与制造: 1-7[2022-02-27]. https://kns.cnki.net/kcms/detail/21.1140.TH.20211114.1517.016.html

[22]舒奇, 黄家才, 林健, 等. 机器人远程控制与自适应抓取研究[J]. 机械设计与制造, 2020(12): 272-276.

[23]阳雨妍, 宋爱国, 沈书馨, 等. 基于CNN-GRU的遥操作机器人操作者识别与自适应速度控制方法[J]. 仪器仪表学报, 2021, 42(03): 123-131.

[24]杨超, 张铭钧, 吴珍臻, 等. 作业型水下机器人纵、横倾姿态自适应区域控制方法[J]. 机器人, 2021, 43(02): 224-233.

[25]蒙双. 基于机器视觉的车道线检测算法研究[D]. 桂林: 桂林电子科技大学, 2020.

[26]Mammeri A, Boukerche A, Tang Z. A real-time lane marking localization, tracking and communication system[J]. Computer Communications, 2016, 73: 132-143.

[27]Zhang X, Chen F. Lane line edge detection based on improved adaptive canny algorithm[C]//Journal of Physics: Conference Series, IOP Publishing, 2020, 1549(2): 022131.

[28]胡晓霞, 李雪梅, 白燕燕, 等. 无人驾驶车道检测技术的研究[J]. 电子设计工程, 2020, 28(18): 118-121+126.

[29]蔡英凤, 高力, 孙晓强, 等. 一种基于形态学特征的车道线识别方法[J]. 中国机械工程, 2018, 29(15): 1863-1868.

[30]Du H. Lane line detection and vehicle identification using monocular camera based on MATLAB[J]. Academic Journal of Computing & Information Science, 2021, 4(2): 55-60.

[31]吴清智. 结构化道路车道线检测方法研究[D]. 海口: 海南大学, 2020.

[32]万斌, 童亮, 魏双叶, 等. 基于自适应滑动窗口检测的车道线识别[J]. 北京信息科技大学学报(自然科学版), 2021, 36(04): 44-49.

[33]魏笑. 基于图像处理的车道线识别[D]. 大连: 大连交通大学, 2020.

[34]王其东, 魏振亚, 王乃汉, 等. 基于投影统计与双曲线拟合的车道识别算法[J]. 中国机械工程, 2019, 30(04): 393-399.

[35]穰玉双. 基于人眼视觉认知原理的车道线识别[D]. 成都: 电子科技大学, 2021.

[36]李茂月, 吕虹毓, 王飞, 等. 基于机器视觉的智能车辆鲁棒车道线识别方法[J]. 中国机械工程, 2021, 32(02): 242-251.

[37]Wang Z, Ren W, Qiu Q. LaneNet: Real-time lane detection networks for autonomous driving[J]. arXiv preprint arXiv: 1807. 01726, 2018.

[38]徐哲钧, 张暐, 郭昊, 等. 雾天车道线识别方法: FoggyCULane数据集的创建[J/OL]. 计算机工程与应用: 1-11[2022-04-19]. http://kns.cnki.net/kcms/detail/11.2127.TP.20210507.1122.002.html

[39]蔡英凤, 张田田, 王海, 等. 基于实例分割和自适应透视变换算法的多车道线检测[J]. 东南大学学报(自然科学版), 2020, 50(04): 775-781.

[40]朱兴柯, 叶飞, 李斌, 等. 变电站巡检机器人运动控制系统研究[J]. 机电信息, 2014(30):122-124.

[41]钱厚亮, 郁汉琪, 解乃军. 基于模糊-PI算法的移动机器人运动控制系统的研究[J]. 机床与液压, 2011, 39(21): 60-62+69.

[42]赵建伟, 项远城, 张冬顺, 等. 基于自适应算法驱动器的矿用机器人运动控制[J]. 煤炭工程, 2018, 50(07): 158-162.

[43]孙棣华, 李硕, 崔明月, 等. 轮式移动机器人智能变结构控制算法研究[J]. 控制工程, 2013, 20(03): 553-557+561.

[44]龚兰芳, 许伦辉. 四旋翼机器人运动控制与自适应PID控制算法设计[J]. 机械设计与制造, 2018(12): 254-256.

[45]韩沛文, 周靖, 蒋林, 等. 焊接机器人运动控制算法研究[J]. 制造业自动化, 2017, 39(07): 152-156.

[46]孙玉山, 李岳明, 张英浩, 等. 改进的模拟退火算法在水下机器人S面运动控制参数优化中的应用[J]. 兵工学报, 2013, 34(11): 1418-1423.

[47]唐旭东, 庞永杰, 李晔. 水下机器人运动的免疫控制方法[J]. 电机与控制学报, 2007(06): 676-680.

[48]王星, 白尚旺, 潘理虎, 等. 一种矿井图像增强算法[J]. 工矿自动化, 2017, 43(3): 48-52.

[49]He K, Sun J, Tang X. Single image haze removal using dark channel prior[J]. IEEE transactions on pattern analysis and machine intelligence, 2010, 33(12): 2341-2353.

[50]张英俊, 雷耀花, 潘理虎. 基于暗原色先验的煤矿井下图像增强技术[J]. 工矿自动化, 2015, 41(03): 80-83.

[51]Land E H, McCann J J. Lightness and retinex theory[J]. Journal of the Optical Society of America, 1971, 61(1): 1-11.

[52]Land E H. The retinex theory of color vision[J]. Scientific American, 1977, 237(6): 108-129.

[53]Jobson D J, Rahman Z, Woodell G A. Properties and performance of a center/surround retinex[J]. IEEE transactions on image processing, 1997, 6(3): 451-462.

[54]Rahman Z, Jobson D J, Woodell G A. Multi-scale retinex for color image enhancement[C]//Proceedings of 3rd IEEE international conference on image processing, IEEE, 1996, 3: 1003-1006.

[55]Jobson D J, Rahman Z, Woodell G A. A multiscale retinex for bridging the gap between color images and the human observation of scenes[J]. IEEE Transactions on Image processing, 1997, 6(7): 965-976.

[56]徐荣鑫, 司垒, 魏英豪, 等. 综采工作面监控图像混合增强算法[J]. 工矿自动化, 2016, 42(04): 36-40.

[57]王焱, 关南楠, 刘海涛. 改进的多尺度Retinex井下图像增强算法[J]. 辽宁工程技术大学学报(自然科学版), 2016, 35(04): 440-443.

[58]阮顺领, 刘丹洋, 白宝军, 等. 基于自适应MSRCP算法的煤矿井下图像增强方法[J]. 矿业研究与开发, 2021, 41(11): 186-192.

[59]范凌云, 梁修荣. 基于小波分解子带直方图匹配的矿井视频图像增强方法[J]. 金属矿山, 2016(06): 130-133.

[60]田子建, 王满利, 吴君, 等. 基于双域分解的矿井下图像增强算法[J]. 光子学报, 2019, 48(05): 107-119.

[61]智宁, 毛善君, 李梅. 基于照度调整的矿井非均匀照度视频图像增强算法[J]. 煤炭学报, 2017, 42(08): 2190-2197.

[62]智宁, 毛善君, 李梅. 基于双伽马函数的煤矿井下低亮度图像增强算法[J]. 辽宁工程技术大学学报(自然科学版), 2018, 37(01): 191-197.

[63]Wang C, Ye Z. Brightness preserving histogram equalization with maximum entropy: a variational perspective[J]. IEEE Transactions on Consumer Electronics, 2005, 51(4): 1326-1334.

[64]Pizer S M, Amburn E P, Austin J D, et al. Adaptive histogram equalization and its variations[J]. Computer vision, graphics, and image processing, 1987, 39(3): 355-368.

[65]Reza A M. Realization of the contrast limited adaptive histogram equalization(CLAHE) for real-time image enhancement[J]. Journal of VLSI signal processing systems for signal, image and video technology, 2004, 38(1): 35-44.

[66]薛国华. 基于综采工作面视频监控的图像增强方法[J]. 煤矿安全, 2021, 52(10): 136-141.

[67]张旭辉, 杜永刚, 霍鑫健, 等. 工作面巡检机器人柔性轨道设计与运动学仿真[J/OL]. 煤炭科学技术: 1-7[2022-04-19]. https://kns.cnki.net/kcms/detail/11.2402.TD.20210512.1825.044.html

[68]Wang W, Yuan X, Chen Z, et al. Weak-light image enhancement method based on adaptive local gamma transform and color compensation[J]. Journal of Sensors, 2021. DOI: https://doi.org/10.1155/2021/5563698

[69]He K, Sun J, Tang X. Guided image filtering[J]. IEEE transactions on pattern analysis and machine intelligence, 2012, 35(6): 1397-1409.

[70]He K, Sun J. Fast guided filter[J]. arXiv preprint arXiv: 1505.00996, 2015.

[71]刘志成, 王殿伟, 刘颖, 等. 基于二维伽马函数的光照不均匀图像自适应校正算法[J]. 北京理工大学学报, 2016, 36(02): 191-196+214.

[72]刘艳, 李腾飞. 对张正友相机标定法的改进研究[J]. 光学技术, 2014, 40(06): 565-570.

[73]宁帮祥. 基于FPGA的图像边缘检测系统设计[D]. 太原: 中北大学, 2021.

[74]林梁豪. 基于图像处理的结构化道路识别算法研究[D]. 哈尔滨: 哈尔滨理工大学, 2021.

[75]赵夫群, 马玉, 戴翀. 基于改进随机抽样一致的点云分割算法[J]. 科学技术与工程, 2021, 21(22): 9455-9460.

[76]黄一哲. 基于S曲线步进式模糊神经网络PID的3D打印精度提升策略[D]. 郑州: 河南大学, 2020.

[77]杨九林, 朱建公, 廖璇, 等. 直流无刷电机S曲线加减速控制算法及其实现[J]. 机床与液压, 2020, 48(23): 160-165.

[78]张旭辉, 刘博兴, 张超, 等. 基于模糊PID的智能掘进机器人系统轨迹纠偏[J]. 煤矿机械, 2021, 42(05): 194-197.

[79]颜廷玉. 直流无刷电机控制系统的设计与研究[D]. 哈尔滨: 哈尔滨工业大学, 2009.