复杂工况下专用焊接机器人运动控制的研究
|
摘要
在复杂焊接工况下焊接机器人是否具有较高的运动控制精度是实现精准焊缝跟踪的关键。本文研究了一种以提高专用焊接机器人运动控制精度为目的,以PC+运动控制卡为控制硬件基础,以模糊神经网络控制系统为控制软件核心,在复杂工况下能实时焊缝跟踪,能满足自动化焊接要求的焊接机器人运动控制系统。
论文研究了一种依靠摆动式电弧传感器来进行实时焊缝跟踪的方式,分析了这种方式原理上的可行性,应用上较其他方式的优越性。在分析总结模糊控制和神经网络控制方法的基础上,提出了针对两者间优缺点互补的模糊神经网络控制方法,从而形成了一套完整的焊缝跟踪及控制模式。
论文对五自由度直角坐标机器人进行了运动学、动力学逆问题的分析,研究了机器人的动态特性,为运动控制系统的研究打下了基础。在此基础上,设计了模糊神经网络控制系统的总体结构,研究了该控制系统的两个核心问题:模糊神经控制器和机器人运动状态辨识器。针对控制器和辨识器中都用到的经典BP神经算法的不足,提出了一种性能更佳的改进BP神经算法。最后运用MATLAB的Simulink仿真软件包对控制器、机器人运动状态辨识器以及整个控制系统的性能进行了仿真分析。
论文将研究的成果运用到工程实际中,设计了控制系统上位机、下位机和它们之间串口通信模块的软、硬件。焊接机器人系统经调试后,进行了整体性能验证试验,取得了良好的试验效果,达到了课题预期的目的。
Whether the robot has high precision motion control under complex conditions is the key to achieve accurate welding seam tracking.With the purpose to improve the accuracy of the special-purpose welding robot motion control, the motion control system’s hardware based on PC and Motion Control Card, it’s software based on fuzzy neural network control system, it could implemented real-time seam tracking and auto welding under complex conditions.
A method of real-time seam tracking based on a swing style arc sensor was discussed in this thesis. The feasibility of the method was ananlysised, it was considered better than other welding seam tracking methods. The methods of fuzzy control and neural network control was analyzed and summarized at first, than fuzzy neural network control method was put forward to make up for the shortcomings of each other. So a complete set of seam tracking and control pattern was formed.
Kinematics, dynamics inverse problems of the five degrees of freedom rectangular coordinates robot were analysised in this thesis, it the foundation of the design of motion control system. Then the overall structure of the fuzzy neural network control was designed, Two core issues of the control system were studyed—fuzzy neural controller and movement state identifier of the robot. Since the classical back propagation neural algorithm which were used in controller and identifier were inadequate, so a better improvement in the performance of back propagation neural algorithm was proposed. Last, simulink simulation package of MATLAB was used to simulate and analyse the performance of fuzzy neural controller, movement state identifier and the whole control system.
Earlier results of the study were used to engineering practice, software and hardware of the control system’s upper and lower machine and serial communication between them was designed. After debugging, welding experiment of the welding robot system was done, which had got good experimental results and reached the objective of the study subjects.
论文研究了一种依靠摆动式电弧传感器来进行实时焊缝跟踪的方式,分析了这种方式原理上的可行性,应用上较其他方式的优越性。在分析总结模糊控制和神经网络控制方法的基础上,提出了针对两者间优缺点互补的模糊神经网络控制方法,从而形成了一套完整的焊缝跟踪及控制模式。
论文对五自由度直角坐标机器人进行了运动学、动力学逆问题的分析,研究了机器人的动态特性,为运动控制系统的研究打下了基础。在此基础上,设计了模糊神经网络控制系统的总体结构,研究了该控制系统的两个核心问题:模糊神经控制器和机器人运动状态辨识器。针对控制器和辨识器中都用到的经典BP神经算法的不足,提出了一种性能更佳的改进BP神经算法。最后运用MATLAB的Simulink仿真软件包对控制器、机器人运动状态辨识器以及整个控制系统的性能进行了仿真分析。
论文将研究的成果运用到工程实际中,设计了控制系统上位机、下位机和它们之间串口通信模块的软、硬件。焊接机器人系统经调试后,进行了整体性能验证试验,取得了良好的试验效果,达到了课题预期的目的。
Whether the robot has high precision motion control under complex conditions is the key to achieve accurate welding seam tracking.With the purpose to improve the accuracy of the special-purpose welding robot motion control, the motion control system’s hardware based on PC and Motion Control Card, it’s software based on fuzzy neural network control system, it could implemented real-time seam tracking and auto welding under complex conditions.
A method of real-time seam tracking based on a swing style arc sensor was discussed in this thesis. The feasibility of the method was ananlysised, it was considered better than other welding seam tracking methods. The methods of fuzzy control and neural network control was analyzed and summarized at first, than fuzzy neural network control method was put forward to make up for the shortcomings of each other. So a complete set of seam tracking and control pattern was formed.
Kinematics, dynamics inverse problems of the five degrees of freedom rectangular coordinates robot were analysised in this thesis, it the foundation of the design of motion control system. Then the overall structure of the fuzzy neural network control was designed, Two core issues of the control system were studyed—fuzzy neural controller and movement state identifier of the robot. Since the classical back propagation neural algorithm which were used in controller and identifier were inadequate, so a better improvement in the performance of back propagation neural algorithm was proposed. Last, simulink simulation package of MATLAB was used to simulate and analyse the performance of fuzzy neural controller, movement state identifier and the whole control system.
Earlier results of the study were used to engineering practice, software and hardware of the control system’s upper and lower machine and serial communication between them was designed. After debugging, welding experiment of the welding robot system was done, which had got good experimental results and reached the objective of the study subjects.
引文
[1]周远清.智能机器人系统.北京:清华大学出版社.1989:15-16
[2]孙明如.日本焊接自动化和焊接机器人发发展动态.工程机械,2000(1):31-32
[3]刘大胜,李庆杰,曲道奎.焊接机器人的发展现状与趋势.机械工人(热加工),2001(9):6-7
[4] John P.H. Steele, Chris Mnich,Chris Debrunner et al. Development of closed-loop control of robotic welding processes. Industrial Robot, 2005, 32 (4) :350–355
[5] Wan Dong, Hui Li. Off-line programming of Spot-weld Robot for Car-body in White Based on Robcad. International Conference on Mechatronics and Automation,2007,5:22-39
[6] Ji-Hyoung Lee ,Jae-Kwon Kim, Hyung-Shik Kim et al. Development of multi-axis gantry type welding robot system using a pc-based controller. International Symposium on Industrial Electronics,2001,3:1536-1541
[7]吴琳,谭营,唐建.运动控制技术发展与展望.机床与液压,2007,35(7):231-233
[8] Kunpeng He, Hua Sun, Wanjuan Cheng. Application of fuzzy neural network based on T-S model for mobile robot to avoid obstacles. Proceedings of the World Congress on Intelligent Control and Automation,2008:8277-8281
[9] Li Zhijun,Tao Pey Yuen,Ge Shuzhi Sam et al. Robust adaptive control of cooperating mobile manipulators with relative motion. Transactions on Systems, Man, and Cybernetics, Part B,2009,39(1):103-116
[10] Yan Mu-Tian,Cheng Tun-Hua. High accuracy motion control of linear motor drive wire-EDMmachines.International Journal of Advanced Manufacturing Technology,2009,40(9-10):918-928
[11] Zhao D,Li S.Gao, F.Zhu Q. Robust adaptive terminal sliding mode-based synchronised position control formultiple motion axes systems.IET Control Theory and Applications,2009,3(1):136-150
[12] Arimoto S,Kawamura S,Miyazaki F. Bettering operation of robots by learning .Journal of Robotic Systems, 1984, 1 (2) :123-140
[13]石振林.油井珩架焊接机器人视觉焊缝跟踪研究:[硕士学位论文] .武汉:华中科技大学图书馆,2007
[14] Freund E. Nonlinear control with arbitrary pole-placement for industrial robots and manipulators .Int J Robotics Research, 1982, 1(1) :76-78
[15] Xu Peiquan,Xu Guoxiang,Tang Xinhua. A visual seam tracking system for robotic arcwelding. International Journal of Advanced Manufacturing Technology,2008,37(1-2):70-75
[16] Bae Kang-Yul,Park Jin-Hyun. A study on development of inductive sensor for automatic weld seam tracking. Journal of Materials Processing Technology,2006,176(1-3):111-116
[17] You B-H,Kim J-W. A study on an automatic seam tracking system by using an electromagnetic sensor for sheet metal arc welding of butt joints. Proceedings of the Institution of Mechanical Engineers,2002,216(6):911-92
[18]王其隆.弧焊过程质量实时传感与控制.北京:机械工业出版社,2000:62-65
[19] Moon Hyeong-Soon,Ko Sung-Hoon,Kim Jeong-Cheol. Automatic seam tracking in pipeline welding with narrow groove. International Journal of Advanced Manufacturing Technology,2009,41(3-4):234-241
[20]林尚扬.焊接机器人及其应用.北京:机械工业出版社,2000:45-47
[21]姚河清,刘波,李春燕等.一种短路过渡电弧传感器的研制.焊接学报,2006,27(7):85-88
[22]陆跟成.自适应模糊控制电弧传感器设计:[硕士学位论文].南京:河海大学图书馆,2004
[23] Chiou C.B,Chiou C.H,Chu C.M et al. The application of fuzzy control on energy saving for multi-unit room air-conditioners. Applied Thermal Engineering,2009,29(2-3):310-316
[24] Wu, Peitsang. Neural networks and fuzzy control with applications to textile manufacturing and management. Dissertation Abstracts International,58(3):1509
[26]李永久.焊接教学机器人研究:[硕士学位论文].武汉:华中科技大学图书,2008
[25] Yousef Hasan A, Wahba Manal A. Adaptive fuzzy mimo control of induction motors. Expert Systems with Applications,2009,36(3):4171-4175
[27]王士同.模糊系统、模糊神经网络及应用程序设计.上海:上海科学技术文献出版社,1997:202-205
[28]王素芬.模糊系统与神经网络结合的现状.网络与信息,2007,5:69
[29]丛爽.神经网络、模糊系统及其在运动控制中的应用.合肥:中国科技大学出版社,2001
[30]熊有伦.机器人学.北京:机械工业出版社,1993:108-109
[31]王耀南.机器人智能控制工程.北京:科学出版社,2004:234-236
[32]郑建荣. ADAMS虚拟样机技术入门与提高.北京:机械工业出版社,2002:1-5
[33]韩霞.模糊神经网络控制器的优化设计:[硕士学位论文].西安:西安理工大学图书馆,2004
[34] Abdullah S Alwadie.Analysis design and tuning of the two-input two-output fuzzy control systems using the simplified fuzzy rules.Doctor of philosophy,Wayne state university.2003,48(6):2533
[35] Milacic Milos V. Neuro-fuzzy control of weld penetration in laser welding by monitoring diverse signals. Dissertation Abstracts International,2000,61(11):6100
[36] Wai Rong-Jong,Chen Po-Chen. Intelligent tracking control for robot manipulator including actuator dynamics via TSK-type fuzzy neural network. IEEE Transactions on Fuzzy Systems,2004,12(4):552-559
[37]韩力群.人工神经网络理论、设计及应用.北京:化学工业出版社,2007:98-99
[38]杜刚.改进的BP神经网络在地下水质评价中的应用:[硕士学位论文].上海:上海师范大学图书馆,2007
[39]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计.北京:清华大学出版社,2004:65-68
[40] Shiuh-Jer Huang,Ruey-Jing Lian.A Hybrid Fuzzy Logic and Neural Network Algorithm for Robot Motion Control. IEEE transactons on industrial electronics ,1997,44(3):408-417
[41] Tholudur Arun Narayanaswamy. Neural network modeling and optimization of biotechnology processes. Dissertation Abstracts International,1993,60(1):282
[42] Kessler Brian Scott. Development of an integrated approach combining artificial neural network material based modeling with finite element analysis of forming processes. Dissertation Abstracts International,2005,67(1):498
[43]孙炜.智能神经网络的机器人控制理论方法研究:[博士学位论文].湖南:湖南大学图书馆,2002
[44]石辛民、郝整清.模糊控制及其MATLAB仿真.北京:清华大学出版社,2008:47
[45]陈学军.步进电机闭环控制的研究与探讨.中国市场,2006(14):31-32
[46]李江全、张丽、红蕾.Visual Basic串口通信与测控应用技术实战详解.北京:人民邮电出版社,2007:116-117
[47]贾东耀,彭树林,刘华友.Matlab与Visual Basic应用接口设计.电脑与信息技术,2007,15 (1):12-14
[48]谭炎,张凌燕.MATLAB与VB混合编程技术研究.微型计算机,2006,22(5):247-249
[49]腊森,李绍斌,N.R.布拉顿.焊缝几何形状参数的预测和选择.华中工学院报,1986,14(4),569-575
[50]徐初雄.焊接技术问答.北京:机械工业出版社,2000:208-210
[2]孙明如.日本焊接自动化和焊接机器人发发展动态.工程机械,2000(1):31-32
[3]刘大胜,李庆杰,曲道奎.焊接机器人的发展现状与趋势.机械工人(热加工),2001(9):6-7
[4] John P.H. Steele, Chris Mnich,Chris Debrunner et al. Development of closed-loop control of robotic welding processes. Industrial Robot, 2005, 32 (4) :350–355
[5] Wan Dong, Hui Li. Off-line programming of Spot-weld Robot for Car-body in White Based on Robcad. International Conference on Mechatronics and Automation,2007,5:22-39
[6] Ji-Hyoung Lee ,Jae-Kwon Kim, Hyung-Shik Kim et al. Development of multi-axis gantry type welding robot system using a pc-based controller. International Symposium on Industrial Electronics,2001,3:1536-1541
[7]吴琳,谭营,唐建.运动控制技术发展与展望.机床与液压,2007,35(7):231-233
[8] Kunpeng He, Hua Sun, Wanjuan Cheng. Application of fuzzy neural network based on T-S model for mobile robot to avoid obstacles. Proceedings of the World Congress on Intelligent Control and Automation,2008:8277-8281
[9] Li Zhijun,Tao Pey Yuen,Ge Shuzhi Sam et al. Robust adaptive control of cooperating mobile manipulators with relative motion. Transactions on Systems, Man, and Cybernetics, Part B,2009,39(1):103-116
[10] Yan Mu-Tian,Cheng Tun-Hua. High accuracy motion control of linear motor drive wire-EDMmachines.International Journal of Advanced Manufacturing Technology,2009,40(9-10):918-928
[11] Zhao D,Li S.Gao, F.Zhu Q. Robust adaptive terminal sliding mode-based synchronised position control formultiple motion axes systems.IET Control Theory and Applications,2009,3(1):136-150
[12] Arimoto S,Kawamura S,Miyazaki F. Bettering operation of robots by learning .Journal of Robotic Systems, 1984, 1 (2) :123-140
[13]石振林.油井珩架焊接机器人视觉焊缝跟踪研究:[硕士学位论文] .武汉:华中科技大学图书馆,2007
[14] Freund E. Nonlinear control with arbitrary pole-placement for industrial robots and manipulators .Int J Robotics Research, 1982, 1(1) :76-78
[15] Xu Peiquan,Xu Guoxiang,Tang Xinhua. A visual seam tracking system for robotic arcwelding. International Journal of Advanced Manufacturing Technology,2008,37(1-2):70-75
[16] Bae Kang-Yul,Park Jin-Hyun. A study on development of inductive sensor for automatic weld seam tracking. Journal of Materials Processing Technology,2006,176(1-3):111-116
[17] You B-H,Kim J-W. A study on an automatic seam tracking system by using an electromagnetic sensor for sheet metal arc welding of butt joints. Proceedings of the Institution of Mechanical Engineers,2002,216(6):911-92
[18]王其隆.弧焊过程质量实时传感与控制.北京:机械工业出版社,2000:62-65
[19] Moon Hyeong-Soon,Ko Sung-Hoon,Kim Jeong-Cheol. Automatic seam tracking in pipeline welding with narrow groove. International Journal of Advanced Manufacturing Technology,2009,41(3-4):234-241
[20]林尚扬.焊接机器人及其应用.北京:机械工业出版社,2000:45-47
[21]姚河清,刘波,李春燕等.一种短路过渡电弧传感器的研制.焊接学报,2006,27(7):85-88
[22]陆跟成.自适应模糊控制电弧传感器设计:[硕士学位论文].南京:河海大学图书馆,2004
[23] Chiou C.B,Chiou C.H,Chu C.M et al. The application of fuzzy control on energy saving for multi-unit room air-conditioners. Applied Thermal Engineering,2009,29(2-3):310-316
[24] Wu, Peitsang. Neural networks and fuzzy control with applications to textile manufacturing and management. Dissertation Abstracts International,58(3):1509
[26]李永久.焊接教学机器人研究:[硕士学位论文].武汉:华中科技大学图书,2008
[25] Yousef Hasan A, Wahba Manal A. Adaptive fuzzy mimo control of induction motors. Expert Systems with Applications,2009,36(3):4171-4175
[27]王士同.模糊系统、模糊神经网络及应用程序设计.上海:上海科学技术文献出版社,1997:202-205
[28]王素芬.模糊系统与神经网络结合的现状.网络与信息,2007,5:69
[29]丛爽.神经网络、模糊系统及其在运动控制中的应用.合肥:中国科技大学出版社,2001
[30]熊有伦.机器人学.北京:机械工业出版社,1993:108-109
[31]王耀南.机器人智能控制工程.北京:科学出版社,2004:234-236
[32]郑建荣. ADAMS虚拟样机技术入门与提高.北京:机械工业出版社,2002:1-5
[33]韩霞.模糊神经网络控制器的优化设计:[硕士学位论文].西安:西安理工大学图书馆,2004
[34] Abdullah S Alwadie.Analysis design and tuning of the two-input two-output fuzzy control systems using the simplified fuzzy rules.Doctor of philosophy,Wayne state university.2003,48(6):2533
[35] Milacic Milos V. Neuro-fuzzy control of weld penetration in laser welding by monitoring diverse signals. Dissertation Abstracts International,2000,61(11):6100
[36] Wai Rong-Jong,Chen Po-Chen. Intelligent tracking control for robot manipulator including actuator dynamics via TSK-type fuzzy neural network. IEEE Transactions on Fuzzy Systems,2004,12(4):552-559
[37]韩力群.人工神经网络理论、设计及应用.北京:化学工业出版社,2007:98-99
[38]杜刚.改进的BP神经网络在地下水质评价中的应用:[硕士学位论文].上海:上海师范大学图书馆,2007
[39]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计.北京:清华大学出版社,2004:65-68
[40] Shiuh-Jer Huang,Ruey-Jing Lian.A Hybrid Fuzzy Logic and Neural Network Algorithm for Robot Motion Control. IEEE transactons on industrial electronics ,1997,44(3):408-417
[41] Tholudur Arun Narayanaswamy. Neural network modeling and optimization of biotechnology processes. Dissertation Abstracts International,1993,60(1):282
[42] Kessler Brian Scott. Development of an integrated approach combining artificial neural network material based modeling with finite element analysis of forming processes. Dissertation Abstracts International,2005,67(1):498
[43]孙炜.智能神经网络的机器人控制理论方法研究:[博士学位论文].湖南:湖南大学图书馆,2002
[44]石辛民、郝整清.模糊控制及其MATLAB仿真.北京:清华大学出版社,2008:47
[45]陈学军.步进电机闭环控制的研究与探讨.中国市场,2006(14):31-32
[46]李江全、张丽、红蕾.Visual Basic串口通信与测控应用技术实战详解.北京:人民邮电出版社,2007:116-117
[47]贾东耀,彭树林,刘华友.Matlab与Visual Basic应用接口设计.电脑与信息技术,2007,15 (1):12-14
[48]谭炎,张凌燕.MATLAB与VB混合编程技术研究.微型计算机,2006,22(5):247-249
[49]腊森,李绍斌,N.R.布拉顿.焊缝几何形状参数的预测和选择.华中工学院报,1986,14(4),569-575
[50]徐初雄.焊接技术问答.北京:机械工业出版社,2000:208-210