掘进机自动导向姿态检测系统设计与实现
摘要
随着社会经济的不断发展,科技的进步和人民生活水平的不断提高,要求巷道施工向安全、优质、高效、无人化方向发展,加强掘进机向智能化方向发展是一个大的趋势,开发研制自动掘进机是解决该问题的必然方向,掘进机位置及姿态的自动测定是全自动掘进机必须解决的核心问题之一。为实现掘进机的自动导向,需要对掘进机的姿态角(水平角、滚动角、俯仰角)以及水平偏向位移进行实时的测量。
论文研究了激光光斑检测定位系统,通过该系统的光斑定位与跟踪算法获取光斑中心坐标,进而转换为掘进机的水平偏向位移,提供给控制系统,结合航姿仪采集的姿态角数据控制掘进机的自动导向。
针对光斑定位提取掘进机姿态参数,论文选取适用于激光光斑识别与跟踪定位的算法,对某些经典算法进行比较,选取适用于本系统的算法,并对其进行优化。运动光斑的识别与跟踪定位算法主要分为预处理、目标识别、形态学滤波和目标跟踪算法四部分。本文对每部分的算法进行了详细分析,预处理部分采用了改进的基于灰度差值的快速中值滤波算法。目标识别则是采用了改进的Otsu阈值分割法,并引入了自适应波门。目标跟踪定位算法主要对重心法和圆拟合法两种算法进行了误差、精度比较分析。此外,还讨论了暗背景和亮背景两种特殊情形下的光斑识别与跟踪算法。另外,考虑到磁场受到干扰时,航姿仪采集水平偏向角不精确,因而提出两束激光提取水平偏向角,并对其进行建模,从而使得水平偏向角的获取更加稳定。
最后,本文通过实验测试验证了算法的可行性,高精度性。各种不同环境下的测试,验证本系统光斑定位的精确性,抗干扰性,满足系统需求。并设计了标定系统对掘进机姿态检测系统采集的姿态角及偏向位移进行标定,进一步验证系统的精确性。
With the continuous development of the socio-economic,advancement of technology and the improvement of people's living standards, require construction of the tunnel to the development of safe、high quality、efficient、unmanned direction. strengthening the boring machine to the intelligent direction is a big trend. It is the inevitable direction to develop automatic boring machine, It is one of the core problem to determine position and attitude for the boring machine automatically. To achieve Boring machine auto-oriented, we need to measure the attitude angle of boring machine(deviation, roll, pitch), and horizontal deflection displacement
This paper researched laser spot detection and location system, we use the spot location and tracking algorithm to obtain the coordinates of the spot center, and then converted to a bias level of the boring machine displacement, provided to the control system.combined with the attitude angle data collected by AHRS instrument,to control boring machine auto-oriented.
Aim to extract boring machine attitude parameters from spot location. This paper selected the laser spot identification and tracking、positioning algorithm. we selected some classical algorithm and optimize it. It is divided into, pre-processing. target recognition, morphological filtering and target tracking algorithm in four parts. In this paper detailed analysis of every part of the algorithm, the pre-processing part gived the fast median filtering algorithm based on gray level difference. Target recognition using improved Otsu Thresholding Method, and the introduction of adaptive gate. Target locating and tracking of gravity and circle fitting methods were compared of error and accuracy. It also discussed two special cases of the dark background and bright background of the spot identification and tracking algorithm. In addition, taking into account the distribution of the magnetic field, deviation angle collection of AHRS instrument is inaccurate, and thus the two laser beams to extract deviation angle is proposed, and its modeling, to make deviation angle be more stable.
Finally, to verify the feasibility of the algorithm, high-precision by further experiments. Under various environmental tests to verify the accuracy of the spot location of the system, interference immunity, to meet this system requirements. I designed a calibration system to calibrate the boring machine attitude and bias displacement extract from the boring machine attitude detection system, to further verify the accuracy of the system.
论文研究了激光光斑检测定位系统,通过该系统的光斑定位与跟踪算法获取光斑中心坐标,进而转换为掘进机的水平偏向位移,提供给控制系统,结合航姿仪采集的姿态角数据控制掘进机的自动导向。
针对光斑定位提取掘进机姿态参数,论文选取适用于激光光斑识别与跟踪定位的算法,对某些经典算法进行比较,选取适用于本系统的算法,并对其进行优化。运动光斑的识别与跟踪定位算法主要分为预处理、目标识别、形态学滤波和目标跟踪算法四部分。本文对每部分的算法进行了详细分析,预处理部分采用了改进的基于灰度差值的快速中值滤波算法。目标识别则是采用了改进的Otsu阈值分割法,并引入了自适应波门。目标跟踪定位算法主要对重心法和圆拟合法两种算法进行了误差、精度比较分析。此外,还讨论了暗背景和亮背景两种特殊情形下的光斑识别与跟踪算法。另外,考虑到磁场受到干扰时,航姿仪采集水平偏向角不精确,因而提出两束激光提取水平偏向角,并对其进行建模,从而使得水平偏向角的获取更加稳定。
最后,本文通过实验测试验证了算法的可行性,高精度性。各种不同环境下的测试,验证本系统光斑定位的精确性,抗干扰性,满足系统需求。并设计了标定系统对掘进机姿态检测系统采集的姿态角及偏向位移进行标定,进一步验证系统的精确性。
With the continuous development of the socio-economic,advancement of technology and the improvement of people's living standards, require construction of the tunnel to the development of safe、high quality、efficient、unmanned direction. strengthening the boring machine to the intelligent direction is a big trend. It is the inevitable direction to develop automatic boring machine, It is one of the core problem to determine position and attitude for the boring machine automatically. To achieve Boring machine auto-oriented, we need to measure the attitude angle of boring machine(deviation, roll, pitch), and horizontal deflection displacement
This paper researched laser spot detection and location system, we use the spot location and tracking algorithm to obtain the coordinates of the spot center, and then converted to a bias level of the boring machine displacement, provided to the control system.combined with the attitude angle data collected by AHRS instrument,to control boring machine auto-oriented.
Aim to extract boring machine attitude parameters from spot location. This paper selected the laser spot identification and tracking、positioning algorithm. we selected some classical algorithm and optimize it. It is divided into, pre-processing. target recognition, morphological filtering and target tracking algorithm in four parts. In this paper detailed analysis of every part of the algorithm, the pre-processing part gived the fast median filtering algorithm based on gray level difference. Target recognition using improved Otsu Thresholding Method, and the introduction of adaptive gate. Target locating and tracking of gravity and circle fitting methods were compared of error and accuracy. It also discussed two special cases of the dark background and bright background of the spot identification and tracking algorithm. In addition, taking into account the distribution of the magnetic field, deviation angle collection of AHRS instrument is inaccurate, and thus the two laser beams to extract deviation angle is proposed, and its modeling, to make deviation angle be more stable.
Finally, to verify the feasibility of the algorithm, high-precision by further experiments. Under various environmental tests to verify the accuracy of the spot location of the system, interference immunity, to meet this system requirements. I designed a calibration system to calibrate the boring machine attitude and bias displacement extract from the boring machine attitude detection system, to further verify the accuracy of the system.
引文
[1]李建英等.臂式掘进机在我国铁路隧道施上中的应用前景.铁道建筑技术,2008,35(2)：4-8..
[2]吴翠艳,黄世功.悬臂式掘进机发展与展望.水力采煤与管道运输,2007,23(4)：45-47.
[3]胡宁.掘进机自动掘进系统的研究与应用[J].重庆：重庆大学,2006.
[4]毛君,吴常田.浅谈悬臂掘进机的发展及趋势[J].中国工程机械学报,2007,5(2)：240-242.
[5]田原.悬臂掘进机自动导向和定位技术探索.煤矿总院太原研究所,2010,15(6)：27-32.
[6]吕国岭,黄威然,庞红军.盾构自动导向方法的应用研究[J].隧道建设.2005,25(5)：10-13.
[7]毛君,李申岩.陀螺仪和全站仪组合的导向系统.辽宁工程大学机械学院,2007.
[8]田原,董爱锋,一种摄像机和激光指向仪相对位姿标定方法.煤矿总院太原研究所,2010,15(6)：34-40..
[9]吴佳梁,梁坚毅.全自动掘进机[p].中国专利：101169038[A],2008-04-30.
[10]唐玄之.激光及其应用.南京农业大学理学院,2002,24(2)：6-9.
[11]蓝章礼,杨小帆.激光光斑中心定位算法的实用性改进.计算机工程,2008,34(06)：7-9.
[12]韩阳.基于数字CMOS图像传感器的激光光斑检测系统[J].天津：天津大学,2010.
[13]刘振高.MRH三维航姿仪的研究.北京：北京机械工业学院,2007.
[14]吴玉莲,王鹏.基于中值的图像椒盐噪声的非迭代滤除[J].电子技术,2008.
[15]Jean-Francois Lapointe. On-Screen Laser Spot Detection for Large Display Interaction. IEEE International Workshop on Haptic Audio Environments,2005.10(1):72-76.
[16]Ying Weiming, Ying Wang. A quick algorithm for filtering noise from laser terrain image[J]. Opt & Laser Tech,2009,2(1):1-13..
[17]Wang Changhong. A novel improved median filter for salt-and-pepper noise from highly corrupted images,2010,3(1):718-722.
[18]Ko S J. Center Weighted Median Filters and Their Applications to Image Enthancement[J]. IEEE Transactions on Circuits and Systems,2005,38(2):984-993.
[19]Restrepo A, Bovik A C.Adaptive Trimmed Mean Filters for Image Restoration[J]. IEEE Transactions on Acoustics, Speech and Signal Processing,2008,36(13):326-337.
[20]杨耿.运动目标的图像识别与跟踪研究[J].南京：江苏大学,2005.
[21]韩思奇,王蕾.图像分割的阈值法综述[J].系统工程与电子技术,2002,24(3)：18-25.
[22]万德寅.一种实用的光斑图像分割算法[J].软件导刊,2012,11(1)：3-5.
[23]韩亚荣,黄朝兵.一种光斑图像的阈值分割和光斑中心坐标的计算方法[J].图形图像,2009,23(3)：16-20.
[24]舒红平,蒋建民.基于灰度最优阈值的图像分割方法及应用[J].重庆工商大学学报,2003,12(2)：45-52.
[25]钟左峰.基于颜色的二维Otsu运动检测算法[J].图形图像,2008,22(2)：34-38.
[26]Kapur JN, Wong AKC.A New Method for Grey Level Picture Thresholding Using the Entropy of the Histogram[J].Computer Vision, Graphics and Image Processing,2006,29(1): 273-285.
[27]Xiangqi Huang, Takeshi Sasaki. Circle Detection and Fitting Based Positioning System Using Laser Range Finder[J].IEEE,2010,3(1):442-447.
[28]杨红丽.目标识别与跟踪在激光主动侦查中的研究[J].成都：电子科技大学,2009.
[29]赵世喆.基于序列图像的运动目标检测和跟踪算法研究[J].北京：北方工业大学,2006.
[30]王薇.光斑图抑噪预处理的方法研究.激光技术,2007,31(1)：55-60.
[31]刘冠兰.几种激光光斑中心检测方法的比较[J].海洋测绘,2007,27(2)：74-76.
[32]王亚丽.基于图像的激光光斑识别和特性研究[J].长春：长春理工大学,2008.
[33]杨耀权,施仁,于希宁.用Hough变换提高激光光斑中心定位精度的算法[J].光学学报,1999,19(12)：1655-1660.
[34]孔兵,王昭,谭玉山.基于圆拟合的激光光斑中心检测算法.红外与激光工程,2002,6(31)：275-279.
[35]Thomas S M, Chan Y T.A simple approach for the estimation of circular arc center and its radius[J].Computer Vision,Graphics,and Image Processing,2006,45(3):362~370.
[36]F.Lagattu. Limits and possibilities of laser speckle and white-light image-correlation methods:theory and experiments. Appl Opt.,2002,41(31):6603-6613.
[37]Liu Ke.Radius constraint least-square circle fitting method and error analysis.IEEE, 2006,17(5):604-607.
[38]Ping S.Huang. Real-time Analysis of Laser Speckle Patterns for Precision Positioning[J]. Signal Processing and Applications,2010,26(2):345-352.
[39]J W Cui, J B Tan. Optimized Algorithm of Laser Spot Center Location in Strong Noise. Journal of Physics:Conference Series,2005,13(2):32-45.
[40]Frosio I, Borghese N.A. Real-time accurate circle fitting with occlusions. IEEE,, 2008,41(3):1041-1055.
[41]Jiaju Ying. Analysis on Laser Spot Locating Precision affected by CMOS Sensor Fill Factor[J].ICEMI,2009,9(1):2202-2206.
[42]H. Tamura, T. Sasaki. Circle Fitting Based Position Measurement System Using Laser Range Finder in Construction Fields.2010 IEEE/RSJ International Conference on Intelligent Robots and Systems,2010,23(1):209-214.
[43]胡林亭,显葵,金俊坤等.CCD测量激光光斑方法研究[J],激光技术,2001,25(2),154-157.
[44]冀伟.激光束位置及角度偏差测量系统[J].长春：长春理工大学,2007.
[45]刘金会,郝静如.自主移动机器人导航定位技术研究初探[J].重庆：重庆大学自动化学研究所,2005.
[2]吴翠艳,黄世功.悬臂式掘进机发展与展望.水力采煤与管道运输,2007,23(4)：45-47.
[3]胡宁.掘进机自动掘进系统的研究与应用[J].重庆：重庆大学,2006.
[4]毛君,吴常田.浅谈悬臂掘进机的发展及趋势[J].中国工程机械学报,2007,5(2)：240-242.
[5]田原.悬臂掘进机自动导向和定位技术探索.煤矿总院太原研究所,2010,15(6)：27-32.
[6]吕国岭,黄威然,庞红军.盾构自动导向方法的应用研究[J].隧道建设.2005,25(5)：10-13.
[7]毛君,李申岩.陀螺仪和全站仪组合的导向系统.辽宁工程大学机械学院,2007.
[8]田原,董爱锋,一种摄像机和激光指向仪相对位姿标定方法.煤矿总院太原研究所,2010,15(6)：34-40..
[9]吴佳梁,梁坚毅.全自动掘进机[p].中国专利：101169038[A],2008-04-30.
[10]唐玄之.激光及其应用.南京农业大学理学院,2002,24(2)：6-9.
[11]蓝章礼,杨小帆.激光光斑中心定位算法的实用性改进.计算机工程,2008,34(06)：7-9.
[12]韩阳.基于数字CMOS图像传感器的激光光斑检测系统[J].天津：天津大学,2010.
[13]刘振高.MRH三维航姿仪的研究.北京：北京机械工业学院,2007.
[14]吴玉莲,王鹏.基于中值的图像椒盐噪声的非迭代滤除[J].电子技术,2008.
[15]Jean-Francois Lapointe. On-Screen Laser Spot Detection for Large Display Interaction. IEEE International Workshop on Haptic Audio Environments,2005.10(1):72-76.
[16]Ying Weiming, Ying Wang. A quick algorithm for filtering noise from laser terrain image[J]. Opt & Laser Tech,2009,2(1):1-13..
[17]Wang Changhong. A novel improved median filter for salt-and-pepper noise from highly corrupted images,2010,3(1):718-722.
[18]Ko S J. Center Weighted Median Filters and Their Applications to Image Enthancement[J]. IEEE Transactions on Circuits and Systems,2005,38(2):984-993.
[19]Restrepo A, Bovik A C.Adaptive Trimmed Mean Filters for Image Restoration[J]. IEEE Transactions on Acoustics, Speech and Signal Processing,2008,36(13):326-337.
[20]杨耿.运动目标的图像识别与跟踪研究[J].南京：江苏大学,2005.
[21]韩思奇,王蕾.图像分割的阈值法综述[J].系统工程与电子技术,2002,24(3)：18-25.
[22]万德寅.一种实用的光斑图像分割算法[J].软件导刊,2012,11(1)：3-5.
[23]韩亚荣,黄朝兵.一种光斑图像的阈值分割和光斑中心坐标的计算方法[J].图形图像,2009,23(3)：16-20.
[24]舒红平,蒋建民.基于灰度最优阈值的图像分割方法及应用[J].重庆工商大学学报,2003,12(2)：45-52.
[25]钟左峰.基于颜色的二维Otsu运动检测算法[J].图形图像,2008,22(2)：34-38.
[26]Kapur JN, Wong AKC.A New Method for Grey Level Picture Thresholding Using the Entropy of the Histogram[J].Computer Vision, Graphics and Image Processing,2006,29(1): 273-285.
[27]Xiangqi Huang, Takeshi Sasaki. Circle Detection and Fitting Based Positioning System Using Laser Range Finder[J].IEEE,2010,3(1):442-447.
[28]杨红丽.目标识别与跟踪在激光主动侦查中的研究[J].成都：电子科技大学,2009.
[29]赵世喆.基于序列图像的运动目标检测和跟踪算法研究[J].北京：北方工业大学,2006.
[30]王薇.光斑图抑噪预处理的方法研究.激光技术,2007,31(1)：55-60.
[31]刘冠兰.几种激光光斑中心检测方法的比较[J].海洋测绘,2007,27(2)：74-76.
[32]王亚丽.基于图像的激光光斑识别和特性研究[J].长春：长春理工大学,2008.
[33]杨耀权,施仁,于希宁.用Hough变换提高激光光斑中心定位精度的算法[J].光学学报,1999,19(12)：1655-1660.
[34]孔兵,王昭,谭玉山.基于圆拟合的激光光斑中心检测算法.红外与激光工程,2002,6(31)：275-279.
[35]Thomas S M, Chan Y T.A simple approach for the estimation of circular arc center and its radius[J].Computer Vision,Graphics,and Image Processing,2006,45(3):362~370.
[36]F.Lagattu. Limits and possibilities of laser speckle and white-light image-correlation methods:theory and experiments. Appl Opt.,2002,41(31):6603-6613.
[37]Liu Ke.Radius constraint least-square circle fitting method and error analysis.IEEE, 2006,17(5):604-607.
[38]Ping S.Huang. Real-time Analysis of Laser Speckle Patterns for Precision Positioning[J]. Signal Processing and Applications,2010,26(2):345-352.
[39]J W Cui, J B Tan. Optimized Algorithm of Laser Spot Center Location in Strong Noise. Journal of Physics:Conference Series,2005,13(2):32-45.
[40]Frosio I, Borghese N.A. Real-time accurate circle fitting with occlusions. IEEE,, 2008,41(3):1041-1055.
[41]Jiaju Ying. Analysis on Laser Spot Locating Precision affected by CMOS Sensor Fill Factor[J].ICEMI,2009,9(1):2202-2206.
[42]H. Tamura, T. Sasaki. Circle Fitting Based Position Measurement System Using Laser Range Finder in Construction Fields.2010 IEEE/RSJ International Conference on Intelligent Robots and Systems,2010,23(1):209-214.
[43]胡林亭,显葵,金俊坤等.CCD测量激光光斑方法研究[J],激光技术,2001,25(2),154-157.
[44]冀伟.激光束位置及角度偏差测量系统[J].长春：长春理工大学,2007.
[45]刘金会,郝静如.自主移动机器人导航定位技术研究初探[J].重庆：重庆大学自动化学研究所,2005.