机器人视觉伺服系统的研究
|
摘要
本文以图像处理算法与滑模变结构控制理论为基础，选取机械手及乒乓球作为研究对象，并且开发了一个用C++语言编写的图像分割技术与伺服控制系统软件。为了实现机器人具有能够根据视觉信息抓取目标的能力，需要从摄像机获取的二维图像中提取出目标特征,但是所获取的图像受到不均匀光线的干扰，导致图像轮廓出现断裂等现象，从而影响到对目标特征识别的效果，进而导致视觉伺服系统在现实的环境中表现不佳等问题。针对这一问题，本文采用了滤波法去除干扰，并且尝试使用支持向量机方法对图像进行分割处理。
支持向量机能够对非线性图像数据进行有效地分类，因此非常适合应用于分割机器人视觉系统中的目标图像。本文采用支持向量机方法对目标图像数据进行分类，实验结果表明：只要适当地选择样本量、样本特征，则该方法就能够有效地分割边界模糊、灰度不均匀等情况下的目标图像，并且克服了依靠大量实验来确定分割阈值的不确定性问题。
针对机械手的建模误差及不确定性问题，本文结合了机器人常用的滑模变结构控制方法，以及分析了变结构中存在的抖振问题，并对机械手的运动轨迹进行了跟踪仿真研究。在此基础上，提出了一种径向基函数神经网络(Radical Basis Function NeuralNetwork)与基于名义模型的滑模变结构控制方法来削弱滑动模态在控制机械手运动中存在的抖振，并利用李亚普诺夫函数方法分析机械手运动的稳定性，仿真结果验证了该方法是可行的。
This paper is based on the image processing algorithm and sliding mode variablestructure control theory, and choose mechanical arm and ping-pong as the research object,working out a software programme for image segmentation technology and servo controlsystem by VC. To realize the robot has able to grab target according to visual information,which is need to extract features of the target from2d image from camera, but the imagewas disturbed from the uneven light, leading to outline of the image appear rupturephenomenon, which influence recognition effect of the target feature, finally, this will leadto poor performance of visual servo system in the real environment. Pointing to theproblem, filtering methods have been input to remove interference and try to use thesupport vector machine method to do image processing in this paper.
SVM (support vector machine) has able to classify nonlinear image data effectively,which is very suitable to segment target image of the robot visual system. This paper usingsupport vector machine classify image data of the target, the experimental results show thatif only choose appropriate sample size and feature, this method has able to segment targetimage with fuzzy boundaries and gray uneven, and overcome uncertainty problems thatrely on extensive experiments to determine segmentation thresholds.
As for the modeling error and uncertain problems in the manipulator, this papercombines the sliding mode variable structure control method that often application forrobot, and analyzes chattering problem which existing in the variable structure, and carryon simulation research to trajectory of the manipulator. The paper extended it into a slidingmode variable structure control method based on the radial Basis Function Neural Networkand the name model to weaken chattering when sliding mode control manipulatormovement and analysis the stability of the manipulator movement by the Lyapunovfunction, and the simulation results verify that the method is feasible.
支持向量机能够对非线性图像数据进行有效地分类，因此非常适合应用于分割机器人视觉系统中的目标图像。本文采用支持向量机方法对目标图像数据进行分类，实验结果表明：只要适当地选择样本量、样本特征，则该方法就能够有效地分割边界模糊、灰度不均匀等情况下的目标图像，并且克服了依靠大量实验来确定分割阈值的不确定性问题。
针对机械手的建模误差及不确定性问题，本文结合了机器人常用的滑模变结构控制方法，以及分析了变结构中存在的抖振问题，并对机械手的运动轨迹进行了跟踪仿真研究。在此基础上，提出了一种径向基函数神经网络(Radical Basis Function NeuralNetwork)与基于名义模型的滑模变结构控制方法来削弱滑动模态在控制机械手运动中存在的抖振，并利用李亚普诺夫函数方法分析机械手运动的稳定性，仿真结果验证了该方法是可行的。
This paper is based on the image processing algorithm and sliding mode variablestructure control theory, and choose mechanical arm and ping-pong as the research object,working out a software programme for image segmentation technology and servo controlsystem by VC. To realize the robot has able to grab target according to visual information,which is need to extract features of the target from2d image from camera, but the imagewas disturbed from the uneven light, leading to outline of the image appear rupturephenomenon, which influence recognition effect of the target feature, finally, this will leadto poor performance of visual servo system in the real environment. Pointing to theproblem, filtering methods have been input to remove interference and try to use thesupport vector machine method to do image processing in this paper.
SVM (support vector machine) has able to classify nonlinear image data effectively,which is very suitable to segment target image of the robot visual system. This paper usingsupport vector machine classify image data of the target, the experimental results show thatif only choose appropriate sample size and feature, this method has able to segment targetimage with fuzzy boundaries and gray uneven, and overcome uncertainty problems thatrely on extensive experiments to determine segmentation thresholds.
As for the modeling error and uncertain problems in the manipulator, this papercombines the sliding mode variable structure control method that often application forrobot, and analyzes chattering problem which existing in the variable structure, and carryon simulation research to trajectory of the manipulator. The paper extended it into a slidingmode variable structure control method based on the radial Basis Function Neural Networkand the name model to weaken chattering when sliding mode control manipulatormovement and analysis the stability of the manipulator movement by the Lyapunovfunction, and the simulation results verify that the method is feasible.
引文
[1]蔡自兴.机器人学(第2版)[M].北京:清华大学出版社,2009
[2]田梦倩,罗翔.机器人视觉伺服系统结构机实现[J].制造业自动化,2005,27(12):40-42
[3]Shirai Y, Inoue H. Guiding. A Robot by Visual Feedback in Assembling Tasks[J].PatternRecognition,1973,(5):99-108
[4]Hill J, Park W T. Real Time Control of a Robot with a Mobile Camera. IndustrialRobots[J].1979,(9):233-246
[5]方勇纯.机器人视觉伺服研究综述[J].智能系统学报,2008,3(2):109-114
[6]赵清杰,连广宇等人.机器人视觉伺服综述[J].控制与决策,2001,(11):849-852
[7]王麟琨,徐德,谭民.机器人视觉伺服研究进展[J].机器人.2004,26(3):277-282
[8]M C Ngugen. Situation-oriented behavior-based stereo vision to gain robustnessadaptation in manipulator control[A]. Pro of SPIE[C]. Massachusetts,1999,3837:90-97
[9]Garcia N, Mamani G, Reinoso O. Visual servo control of industrial robot. IFAC15thTriennial World Congress. Barcelona, Spain,2002
[10]付秋瑜,林清宇,张万成,吴南健.面向实时视觉芯片的高速CMOS图像传感器[J].光学学报,2011,31(8):1-7
[11]张宇腾,杨明,王春香.基于主动视觉的智能车道路跟踪方法[J].上海交通大学学报,2010,44(8):1033-1045
[12]廖强,刘兆东,郭静,杨正华.主动视觉技术在精密测量中的应用研究[J].计算机工程与应用,2009,45(10):218-220
[13]Corke P. Visual control of robot manipulators-A review[J]. Robotic and AutomatedSystems,1993,7:1-31
[14]桑农,张天序,曹治国.基于边缘约束的红外目标图像松弛分割技术[J].电子报,2002,30(7):1027-1031
[15]边肇棋,张学工.模式识别[M].北京:清华大学出版社,2000
[16]林续鹏,刘君华.基于小波的支持向量机算法研究[J].西安交大学报,39(8):816-819
[17]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):33-42
[18]C. Yan，N. Sang and T. Zhang. Local entropy-based transition region extraction andthresholding[J]. Patter Recognition Letters,2003,24:1935-2941
[19]陈寅鹏,丁晓青.复杂车辆图像中的车牌定位与字符分割方法[J].红外与激光上程,2004,33(1):29-33
[20]徐海祥.基于支持向量机方法的图像分割与目标分类.华中科技大学博士论文,2005.6
[21]张鹏,李元春.一种机械臂协调操作柔性负载的振动控制研究[J].系统仿真学报,2010,22(10):2387-2391
[22]唐志国,李元春,刘木林.机械臂协调操作柔性负载鲁棒神经网络控制[J].浙江大学学报(工学版),2010,44(7):1394-1399
[23]刘金琨.先进PID控制Matlab仿真(第2版)[M].北京:电子工业出版社,2004
[24]王昊瀛,王景,吴宏鑫,梁斌.空间机器人的目标捕获自适应控制[J].中国科学空间技术,2000,20(5):1-9
[25]焦晓红,李运锋,方一鸣,耿秋实.一种机器人鲁棒自适应控制法[J].机器人技术与应用,2002,3:40-43
[26]庞中华,崔红.系统辨识与自适应控制Matlab仿真[M].北京:北京航空航天大学出版社,2009
[27]Liu J K, Er L J. QFT Robust Control Design for3-Axis Flight Table Servo Systemwith Large Friction[J].Chinese Journal of Aeronautics,2004,17(1):34-38
[28]李春梅,周骥平,颜景平.人工神经网络在机器人视觉中的应用[J].制造业自动化,2000,22(9):33-36
[29]李二超,李炜.基于BP神经网络的机器人容错控制[J].机器人技术,2007,29(6):24-25
[30]李优新.机器人无标定视觉伺服系统的多种智能控制方法研究.华南理工大学博士论文,2009
[31]王从庆.基于神经网络的机器人迭代学习控制[J].南京航空航天大学学报,1998,30(4):395-399
[32]刘金琨.机器人控制系统的设计与MATLAB仿真[M].北京:清华大学出版社,2008
[33]刘金琨.滑模变结构控制Matlab仿真[M].北京:清华大学出版社,2005
[34]Y N Fei, J S Smith, Q H Wu. Sliding mode control of robot Manipulators based onsliding mode perturbation observation[J]. Systems and Control Engineering,2006,220(1):201-210
[35]Chuan-Kai Lin. Nonsingular terminal sliding mode control of robot manipulators usingfuzzy wavelet networks[J].IEEE tractions on fuzzy systems,2006,14(6):849-859
[36]Chaumette F, Hutchinson S. Visual servo control, Part II, Advanced approaches[J].IEEE Robotics and Automation Magazine,2007,17(3):109-118
[37]张晓晖,刘丁.仿人视觉过程的机器人视觉伺服系统研究[J].计算机工程与应用,2007,43(18):11-15
[38]孙宜标,郭庆鼎.基于RBF神经网络补偿的直线伺服系统滑模鲁棒跟踪控制[J].控制理论与应用,2004,21(2):252-256
[39]马东玺,范大鹏,张连超.数字伺服控制系统软件模块化设计[J].软件技术,2005,24(2):86-88
[40]李昊,傅曦.精通——Visual C++指纹模式识别系统算法及实现[M].北京:人民邮电出版社,2008
[41]张宏林.精通visual c++数字图像处理典型算法及实现(第2版)[M].北京:人民邮电出版社,2008
[42]杨淑莹.图像模式识别:VC++技术实现[M].北京:北方交通大学出版社,2005
[43]杨强.支持向量机的模型及其在图像分割中的应用.重庆大学博士论文,2004
[44]于双和,强文义,傅佩琛.无抖振离散准滑模控制[J].控制与决策,2001,16(3):380-382
[45]张袅娜,冯勇,孙黎霞.双臂柔性机械手的终端滑模控制[J].控制与决策,2004,109(10):1142-1146
[46]Kim S W, Lee J J. Design of a fuzzy controller with fuzzy sliding surface[J]. Fuzzysets and Systems,1995,71(3):359-367
[47]Haiyan Yu, Qing Hu, Xiying Ding. Fuzzy Sliding Mode Variable Structure Control ofThree-Phase Rectifier. Fuzzy Systems and Knowledge Discovery. Fourth InternationalConference on,2007,2:581-585
[48]Cui Jiefan, Wu Hui, Fu Yue. Fuzzy Sliding Mode Variable Structure Control of LinearMotor. Fuzzy Systems and Knowledge Discovery. Sixth International Conference on,2009,4:112-115
[49]Chern T L, Wu Y C. An optimal variable structure controller with integralcompensation for electro hydraulic position servo control systems[J]. IEEETransactions on industrial electronics Proceedings-D,19992,39(5):460-463
[50]Z H Man, X H Yu, K Eshraghian, M Palaniswami. A robot adaptive sliding modetracking control using an RBF neural network for robotic manipulators[J]. IEEEInternational Conference on Neural Networks,1995,5:2403-2408
[51]Corradini M L, Giantomassi A, Ippoliti G, Longhi S, Orlando G. Discrete time variablestructure control of robotic manipulators based on fully tuned RBF neural networks[J].Industrial Electronics,2010IEEE International Symposium on,2010,3:1840-1845
[52]Broomhead D S, Lowe D. Multivariable functional interpolation andadaptivenetworks[J]. Complex Systems,1988,2(2):321-355
[53]IRH JACKSON. An order of convergence for some radial basis functions[J]. IMAJournal of Numerical Analysis,1989,9:567-587
[54]阮晓钢.神经计算科学(在细胞的水平上模拟脑功能)[M].北京:国防工业出版社,2006
[55]徐丽娜.神经网络控制[M].哈尔滨:哈尔滨工业大学出版社.1999
[56]陈海忠,高国琴.两自由度并联机器人的RBF神经网络辨识滑模控制策略研究[J].机床与液压,2011,39(7):21-27
[57]孙鑫,余安萍.VC++深入详解[M].北京:电子工业出版社,2006
[2]田梦倩,罗翔.机器人视觉伺服系统结构机实现[J].制造业自动化,2005,27(12):40-42
[3]Shirai Y, Inoue H. Guiding. A Robot by Visual Feedback in Assembling Tasks[J].PatternRecognition,1973,(5):99-108
[4]Hill J, Park W T. Real Time Control of a Robot with a Mobile Camera. IndustrialRobots[J].1979,(9):233-246
[5]方勇纯.机器人视觉伺服研究综述[J].智能系统学报,2008,3(2):109-114
[6]赵清杰,连广宇等人.机器人视觉伺服综述[J].控制与决策,2001,(11):849-852
[7]王麟琨,徐德,谭民.机器人视觉伺服研究进展[J].机器人.2004,26(3):277-282
[8]M C Ngugen. Situation-oriented behavior-based stereo vision to gain robustnessadaptation in manipulator control[A]. Pro of SPIE[C]. Massachusetts,1999,3837:90-97
[9]Garcia N, Mamani G, Reinoso O. Visual servo control of industrial robot. IFAC15thTriennial World Congress. Barcelona, Spain,2002
[10]付秋瑜,林清宇,张万成,吴南健.面向实时视觉芯片的高速CMOS图像传感器[J].光学学报,2011,31(8):1-7
[11]张宇腾,杨明,王春香.基于主动视觉的智能车道路跟踪方法[J].上海交通大学学报,2010,44(8):1033-1045
[12]廖强,刘兆东,郭静,杨正华.主动视觉技术在精密测量中的应用研究[J].计算机工程与应用,2009,45(10):218-220
[13]Corke P. Visual control of robot manipulators-A review[J]. Robotic and AutomatedSystems,1993,7:1-31
[14]桑农,张天序,曹治国.基于边缘约束的红外目标图像松弛分割技术[J].电子报,2002,30(7):1027-1031
[15]边肇棋,张学工.模式识别[M].北京:清华大学出版社,2000
[16]林续鹏,刘君华.基于小波的支持向量机算法研究[J].西安交大学报,39(8):816-819
[17]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):33-42
[18]C. Yan，N. Sang and T. Zhang. Local entropy-based transition region extraction andthresholding[J]. Patter Recognition Letters,2003,24:1935-2941
[19]陈寅鹏,丁晓青.复杂车辆图像中的车牌定位与字符分割方法[J].红外与激光上程,2004,33(1):29-33
[20]徐海祥.基于支持向量机方法的图像分割与目标分类.华中科技大学博士论文,2005.6
[21]张鹏,李元春.一种机械臂协调操作柔性负载的振动控制研究[J].系统仿真学报,2010,22(10):2387-2391
[22]唐志国,李元春,刘木林.机械臂协调操作柔性负载鲁棒神经网络控制[J].浙江大学学报(工学版),2010,44(7):1394-1399
[23]刘金琨.先进PID控制Matlab仿真(第2版)[M].北京:电子工业出版社,2004
[24]王昊瀛,王景,吴宏鑫,梁斌.空间机器人的目标捕获自适应控制[J].中国科学空间技术,2000,20(5):1-9
[25]焦晓红,李运锋,方一鸣,耿秋实.一种机器人鲁棒自适应控制法[J].机器人技术与应用,2002,3:40-43
[26]庞中华,崔红.系统辨识与自适应控制Matlab仿真[M].北京:北京航空航天大学出版社,2009
[27]Liu J K, Er L J. QFT Robust Control Design for3-Axis Flight Table Servo Systemwith Large Friction[J].Chinese Journal of Aeronautics,2004,17(1):34-38
[28]李春梅,周骥平,颜景平.人工神经网络在机器人视觉中的应用[J].制造业自动化,2000,22(9):33-36
[29]李二超,李炜.基于BP神经网络的机器人容错控制[J].机器人技术,2007,29(6):24-25
[30]李优新.机器人无标定视觉伺服系统的多种智能控制方法研究.华南理工大学博士论文,2009
[31]王从庆.基于神经网络的机器人迭代学习控制[J].南京航空航天大学学报,1998,30(4):395-399
[32]刘金琨.机器人控制系统的设计与MATLAB仿真[M].北京:清华大学出版社,2008
[33]刘金琨.滑模变结构控制Matlab仿真[M].北京:清华大学出版社,2005
[34]Y N Fei, J S Smith, Q H Wu. Sliding mode control of robot Manipulators based onsliding mode perturbation observation[J]. Systems and Control Engineering,2006,220(1):201-210
[35]Chuan-Kai Lin. Nonsingular terminal sliding mode control of robot manipulators usingfuzzy wavelet networks[J].IEEE tractions on fuzzy systems,2006,14(6):849-859
[36]Chaumette F, Hutchinson S. Visual servo control, Part II, Advanced approaches[J].IEEE Robotics and Automation Magazine,2007,17(3):109-118
[37]张晓晖,刘丁.仿人视觉过程的机器人视觉伺服系统研究[J].计算机工程与应用,2007,43(18):11-15
[38]孙宜标,郭庆鼎.基于RBF神经网络补偿的直线伺服系统滑模鲁棒跟踪控制[J].控制理论与应用,2004,21(2):252-256
[39]马东玺,范大鹏,张连超.数字伺服控制系统软件模块化设计[J].软件技术,2005,24(2):86-88
[40]李昊,傅曦.精通——Visual C++指纹模式识别系统算法及实现[M].北京:人民邮电出版社,2008
[41]张宏林.精通visual c++数字图像处理典型算法及实现(第2版)[M].北京:人民邮电出版社,2008
[42]杨淑莹.图像模式识别:VC++技术实现[M].北京:北方交通大学出版社,2005
[43]杨强.支持向量机的模型及其在图像分割中的应用.重庆大学博士论文,2004
[44]于双和,强文义,傅佩琛.无抖振离散准滑模控制[J].控制与决策,2001,16(3):380-382
[45]张袅娜,冯勇,孙黎霞.双臂柔性机械手的终端滑模控制[J].控制与决策,2004,109(10):1142-1146
[46]Kim S W, Lee J J. Design of a fuzzy controller with fuzzy sliding surface[J]. Fuzzysets and Systems,1995,71(3):359-367
[47]Haiyan Yu, Qing Hu, Xiying Ding. Fuzzy Sliding Mode Variable Structure Control ofThree-Phase Rectifier. Fuzzy Systems and Knowledge Discovery. Fourth InternationalConference on,2007,2:581-585
[48]Cui Jiefan, Wu Hui, Fu Yue. Fuzzy Sliding Mode Variable Structure Control of LinearMotor. Fuzzy Systems and Knowledge Discovery. Sixth International Conference on,2009,4:112-115
[49]Chern T L, Wu Y C. An optimal variable structure controller with integralcompensation for electro hydraulic position servo control systems[J]. IEEETransactions on industrial electronics Proceedings-D,19992,39(5):460-463
[50]Z H Man, X H Yu, K Eshraghian, M Palaniswami. A robot adaptive sliding modetracking control using an RBF neural network for robotic manipulators[J]. IEEEInternational Conference on Neural Networks,1995,5:2403-2408
[51]Corradini M L, Giantomassi A, Ippoliti G, Longhi S, Orlando G. Discrete time variablestructure control of robotic manipulators based on fully tuned RBF neural networks[J].Industrial Electronics,2010IEEE International Symposium on,2010,3:1840-1845
[52]Broomhead D S, Lowe D. Multivariable functional interpolation andadaptivenetworks[J]. Complex Systems,1988,2(2):321-355
[53]IRH JACKSON. An order of convergence for some radial basis functions[J]. IMAJournal of Numerical Analysis,1989,9:567-587
[54]阮晓钢.神经计算科学(在细胞的水平上模拟脑功能)[M].北京:国防工业出版社,2006
[55]徐丽娜.神经网络控制[M].哈尔滨:哈尔滨工业大学出版社.1999
[56]陈海忠,高国琴.两自由度并联机器人的RBF神经网络辨识滑模控制策略研究[J].机床与液压,2011,39(7):21-27
[57]孙鑫,余安萍.VC++深入详解[M].北京:电子工业出版社,2006