基于平衡杆控制的刚柔耦合走钢丝机器人的研究
|
摘要
走钢丝机器人是一种能够保持自平衡并行走于拉紧的钢丝之上的机械系统。该类机器人的命名源自于人类的高空走钢丝运动,目的在于揭示走钢丝运动的内在规律,并能够在机械装置上实现这种稳定平衡运动方式。本文研究的基于平衡杆控制的走钢丝机器人与利用飞轮陀螺效应的控制方式有着本质的不同。这种基于平衡杆的控制方式更接近于现实生活中的走钢丝运动。目前针对这类控制方式的走钢丝机器人的研究还不多,已知的研究也是理论性的分析探讨,还没有文献报道过该类机器人样机的实验。
本论文首先对走钢丝运动进行了初步探讨,建立了以纯转动控制平衡杆来实现系统稳定平衡的拉格朗日动力学模型。并基于部分反馈线性化方法设计了控制策略,将电机的输出力矩作为系统输入,选取机器人横滚角度及其角速度、转杆相对机器人本体的转动角度及其角速度作为系统的四个输出变量,建立了系统的控制器。
通过Adams软件对机器人的平衡运动进行了计算机仿真,结果验证了该控制器的有效性。文章进一步设计了该类机器人的物理样机,并搭建了以DSP28335为核心处理器的测控系统,配置了MTi陀螺仪和光电码盘等传感器。样机的实验结果表明,以转动控制可以实现机器人在刚性导轨上的稳定平衡。
考虑到走钢丝运动中,手臂对平衡杆的操作既有转动控制又有左右平动控制,于是改进了机器人结构,利用转动耦合平动来控制。采用相同分析方法,基于新机器人的运动仿真控制和物理样机实验验证了控制器的有效性,以及自平衡收敛速度和抗干扰能力的提高。
高性能的机器人样机为柔性钢丝实验提供了平台,而在柔性钢丝底座上的实验恰好进一步验证了控制器的性能,同时也完成了该系统的刚柔耦合控制实验。
本论文关于走钢丝机器人的研究沿循了从简入手,逐步深化的研究方式。完成了两种结构的走钢丝机器人的分析和样机实验,丰富了基于平衡杆控制的走钢丝机器人领域的内涵。
The wire-walking robot is a kind of mechanical system which can keep self-balance and walk on a tight rope. This kind of robot is named of the high-wire walking show of human people. The intention of the study is to discover the internal law of the sport, and realize this kind of control on the mechanical device. The research of his article is to control the balancing pole to keep the robot self-balance, and this method is totally different from the gyroscopic effect of a flywheel. The former is much more similar to the actual sport. At present, there is not much research on these kind of control method, and only some theoretical analysis. None of the documents had done any experiments on a real robot.
Firstly, some tentative analysis of the wire-walking sport has been done, and the lagrange model of the system that is only controlled by a rolling pole is established. Then the controller of the system is designed under the partial feedback linearization control algorithm. In which, the torque of the motor is taken as the input of the system, and the roll angle and its velocity of the body were taken as two of the system outputs. The other two outputs were the angle and its velocity of the balancing pole relative to the body.
The computer simulation of the balance control is produced in the Adams software, and the simulation results just testified the effectiveness of the controller designed. Then a well-designed robot is made. The measurement and control system of the robot is based on the DSP28335, and the MCU is equipped with a MTi gyroscope and a photoelectricity encoder. The experiment on the robot told that it is available to control the robot walk on a rigid rope steadily only based on a rolling pole.
Considering that people can make the balancing pole roll and move it to their left and right, then the mechanical structure was redesigned. The new robot is controlled by a rolling pole that coupled a translation pole. At the foundation of the former analysis, the control simulation of the new robot also testified the effectiveness of its controller, And the same result is shown with the experiment. The new robot can converge much faster than the old one, and the anti-jamming ability is greater too.
The high performance robot is the precondition to conduct the experiment on the flexible wire. However, the result of the experiment is well enough as the former ones, which means that the rigid-flexible coupling robot is qualified to keep self-balance and walk on a flexible wire.
The research of this article of the wire-walking robot is produced from simple to complex, and finished step by step. The two well-designed robots and the experiments have enriched the knowledge of the wire-walking robot based on the balancing pole.
本论文首先对走钢丝运动进行了初步探讨,建立了以纯转动控制平衡杆来实现系统稳定平衡的拉格朗日动力学模型。并基于部分反馈线性化方法设计了控制策略,将电机的输出力矩作为系统输入,选取机器人横滚角度及其角速度、转杆相对机器人本体的转动角度及其角速度作为系统的四个输出变量,建立了系统的控制器。
通过Adams软件对机器人的平衡运动进行了计算机仿真,结果验证了该控制器的有效性。文章进一步设计了该类机器人的物理样机,并搭建了以DSP28335为核心处理器的测控系统,配置了MTi陀螺仪和光电码盘等传感器。样机的实验结果表明,以转动控制可以实现机器人在刚性导轨上的稳定平衡。
考虑到走钢丝运动中,手臂对平衡杆的操作既有转动控制又有左右平动控制,于是改进了机器人结构,利用转动耦合平动来控制。采用相同分析方法,基于新机器人的运动仿真控制和物理样机实验验证了控制器的有效性,以及自平衡收敛速度和抗干扰能力的提高。
高性能的机器人样机为柔性钢丝实验提供了平台,而在柔性钢丝底座上的实验恰好进一步验证了控制器的性能,同时也完成了该系统的刚柔耦合控制实验。
本论文关于走钢丝机器人的研究沿循了从简入手,逐步深化的研究方式。完成了两种结构的走钢丝机器人的分析和样机实验,丰富了基于平衡杆控制的走钢丝机器人领域的内涵。
The wire-walking robot is a kind of mechanical system which can keep self-balance and walk on a tight rope. This kind of robot is named of the high-wire walking show of human people. The intention of the study is to discover the internal law of the sport, and realize this kind of control on the mechanical device. The research of his article is to control the balancing pole to keep the robot self-balance, and this method is totally different from the gyroscopic effect of a flywheel. The former is much more similar to the actual sport. At present, there is not much research on these kind of control method, and only some theoretical analysis. None of the documents had done any experiments on a real robot.
Firstly, some tentative analysis of the wire-walking sport has been done, and the lagrange model of the system that is only controlled by a rolling pole is established. Then the controller of the system is designed under the partial feedback linearization control algorithm. In which, the torque of the motor is taken as the input of the system, and the roll angle and its velocity of the body were taken as two of the system outputs. The other two outputs were the angle and its velocity of the balancing pole relative to the body.
The computer simulation of the balance control is produced in the Adams software, and the simulation results just testified the effectiveness of the controller designed. Then a well-designed robot is made. The measurement and control system of the robot is based on the DSP28335, and the MCU is equipped with a MTi gyroscope and a photoelectricity encoder. The experiment on the robot told that it is available to control the robot walk on a rigid rope steadily only based on a rolling pole.
Considering that people can make the balancing pole roll and move it to their left and right, then the mechanical structure was redesigned. The new robot is controlled by a rolling pole that coupled a translation pole. At the foundation of the former analysis, the control simulation of the new robot also testified the effectiveness of its controller, And the same result is shown with the experiment. The new robot can converge much faster than the old one, and the anti-jamming ability is greater too.
The high performance robot is the precondition to conduct the experiment on the flexible wire. However, the result of the experiment is well enough as the former ones, which means that the rigid-flexible coupling robot is qualified to keep self-balance and walk on a flexible wire.
The research of this article of the wire-walking robot is produced from simple to complex, and finished step by step. The two well-designed robots and the experiments have enriched the knowledge of the wire-walking robot based on the balancing pole.
引文
[1]晨光.走钢丝机器人.机器人技术与应用,第2期,1995年：17.
[2]周春林,陈诗毅,宋嘉仁,韩天璞,殷跃红.走钢丝机器人[P].上海交通大学,2006.
[3]宋杰文,贾鹏飞,黄文贤,李根.机器人走钢丝的简易模糊控制器[J/OL].中国科技论文在线,http://www.paper.edu.cn.
[4]陈静,阮晓钢,蔡建羡,郜园园.一种欠驱动柔性机器人模型的建立及控制[J].小型微型计算机系统,30(12),2009：2477-2480.
[5]王凯.两轮机器人动力学建模与控制方法研究[D].北京航空航天大学,2007.
[6]阮晓钢,任红格.两轮白平衡机器人动力学建模及其平衡控制.计算机应用研究,第26卷第1期,2009年1月：99-101.
[7]安延涛,马汝建.机械系统的拉格朗日法建模与仿真.系统仿真技术,第3卷第4期,2007年10月：206-210.
[8]Takeshi Tsujimura, Takenori Morimitsu. Dynamics of mobile legs suspended from wire [J]. Robotics and Autonomous Systems, Vol.20, No.1, April 1997: 85-98.
[9]Jun Sawada, Kazuyuki Kusumoto, Yasuhisa Maikawa, Tadashi Munakata, Yoshinobu Ishikawa, A mobile robot for inspection of power transmission lines [J]. IEEE Transactions on Power Delivery, Vol.6, No.1, January,1991:309-315.
[10]T. Tsujimura, T.Yabuta, T. Morimitsu. Design of a wire-suspended mobile robot capable of avoiding path obstacles [J]. IEE Proceedings:Control Theory and Applications, Vol.143, No.4, Jul 1996:349-357.
[11]Ludan Wang, Sheng Cheng, Jianwei Zhang. Development of a line-walking mechanism for power transmission line inspection purpose [A].2009 IEEE/RSJ International Conference on Intelligent Robots and Systems[C]:3323-3328.
[12]Li Tang, Shuangfei Fu, Lijin Fang, Hongguang Wang, Obstacle-navigation strategy of a wire-suspend robot for power transmission lines [A]. Proceedings of the 2004 IEEE International Conference on Robotics and Biomimetics [C]: 82-87.
[13]SunSin Han, JangMyung Lee. Path-selection control of a power line inspection robot using sensor fusion [A]. IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems [C],2008:8-13.
[14]干方建,刘正干.基于传感器的操作臂惯性参数在线识别神经网络模型[J].机械科学与技术,2003,22(1)：168-170.
[15]王林松.基于腕力传感器的机器人操作器动力学参数实时识别[D].合肥工业大学,1999.
[16]王冬,张忠林,王臣业.仿生机器蟹单足惯性参数在线识别神经网络模型[J].机械工程师,2006,(4)：36-38.
[17]李祖枢,张华,古建功,陈桂强.3关节单杠体操机器人的动力学参数辨识[J].控制理论与应用,2008,25(2)：242-246.
[18]工北战,何清华,郝鹏等.挖掘机器人动力学参数辨识研究[J].华中师范大学学报(自然科学版),2009,43(2)：239-242.
[19]黄用华PowerCube模块化机器人惯性参数辨识[D].北京工业大学,2006.
[20]陈恩伟.机器人动态特性及动力学参数辨识研究[D].合肥工业大学,2005.
[21]姜春福.基于神经网络的机器人模型辨识与控制研究[D].北京工业大学,2003.
[22]Zheng Li, Yi Ruan. Obstacle negotiation control for a mobile robot suspended on overhead ground wires by optoelectronic sensors [A]. Proceedings of SPIE-The International Society for Optical Engineering [C], Vol.7508,2009.
[23]Li Tang, Lijin Fang, Hongguang Wang. Obstacle-navigation control for a mobile robot suspended on overhead ground wires [A].8th International Conference on Control, Automation, Robotics and Vision [C], Vol.3,2004:2082-2087.
[24]Y. Takahashi, O. Tsubouchi. Tension control of wire suspended mechanism and application to bathroom cleaning robot [A].39th Annual Conference of the Society of Instrument and Control Engineers [C],2000:143-147.
[25]Noritaka Yanai, Motoji Yamamoto, Akira Mohri. Anti-sway control for wire-suspended mechanism based on dynamics compensation [A]. Proceedings of the 2002 IEEE International Conference on Robotics and Automation [C], Vol.4:4287-4292.
[26]董天立,张超平.走钢丝的力学分析.力学与实践,第4期第27卷,2005年：89-90.
[27]邱秉权.分析力学[M].北京：中国铁道出版社,1998.
[28]陈炜,余跃庆,张绪平.欠驱动机器人研究综述[J].机械设计与研究,21(4)2005：22-26.
[29]高丙团,黄学良.欠驱动2DTORA基于部分反馈线性化的非线性控制设计.东南大学学报,第41卷第2期,2011年3月：321-325.
[30]陈玉敏,曲延滨.倒立摆的局部反馈线性化控制方法.微计算机信息,第25卷第8-1期,2009年：4-6.
[31]胡寿松.自动控制原理[M].科学出版社,2007.
[32]李增刚,Adams入门详解与实例.国防工业大学出版社,2006.
[33]TMS320F28335 Data Manual. Texas Instruments Incorporated,2008.
[34]张玉磊,陈进.基于MTi系统的三维运动检测方法的研究[J].测控技术,2008,27(5)：34-36
[35]黄冀等.MTi微惯性航姿系统/GPS组合技术研究[D].哈尔滨工程大学,2009年2月.
[36]ProMotion BDMC2803/3606/3610直流伺服驱动器用户手册.北京博创兴盛机器人技术有限公司,2010.
[37]徐建柱,刁燕,罗华,高山.基于MATLAB和ADAMS的自平衡机器人联合仿真.现代电子技术,第35卷第6期,2012年3月：90-92.
[38]应再恩,平雪良,陈鲁刚.基于ADAMS和MATLAB的双回路PID控制倒立摆联合仿真.机械传动,第8期第36卷,2012年：64-67.
[2]周春林,陈诗毅,宋嘉仁,韩天璞,殷跃红.走钢丝机器人[P].上海交通大学,2006.
[3]宋杰文,贾鹏飞,黄文贤,李根.机器人走钢丝的简易模糊控制器[J/OL].中国科技论文在线,http://www.paper.edu.cn.
[4]陈静,阮晓钢,蔡建羡,郜园园.一种欠驱动柔性机器人模型的建立及控制[J].小型微型计算机系统,30(12),2009：2477-2480.
[5]王凯.两轮机器人动力学建模与控制方法研究[D].北京航空航天大学,2007.
[6]阮晓钢,任红格.两轮白平衡机器人动力学建模及其平衡控制.计算机应用研究,第26卷第1期,2009年1月：99-101.
[7]安延涛,马汝建.机械系统的拉格朗日法建模与仿真.系统仿真技术,第3卷第4期,2007年10月：206-210.
[8]Takeshi Tsujimura, Takenori Morimitsu. Dynamics of mobile legs suspended from wire [J]. Robotics and Autonomous Systems, Vol.20, No.1, April 1997: 85-98.
[9]Jun Sawada, Kazuyuki Kusumoto, Yasuhisa Maikawa, Tadashi Munakata, Yoshinobu Ishikawa, A mobile robot for inspection of power transmission lines [J]. IEEE Transactions on Power Delivery, Vol.6, No.1, January,1991:309-315.
[10]T. Tsujimura, T.Yabuta, T. Morimitsu. Design of a wire-suspended mobile robot capable of avoiding path obstacles [J]. IEE Proceedings:Control Theory and Applications, Vol.143, No.4, Jul 1996:349-357.
[11]Ludan Wang, Sheng Cheng, Jianwei Zhang. Development of a line-walking mechanism for power transmission line inspection purpose [A].2009 IEEE/RSJ International Conference on Intelligent Robots and Systems[C]:3323-3328.
[12]Li Tang, Shuangfei Fu, Lijin Fang, Hongguang Wang, Obstacle-navigation strategy of a wire-suspend robot for power transmission lines [A]. Proceedings of the 2004 IEEE International Conference on Robotics and Biomimetics [C]: 82-87.
[13]SunSin Han, JangMyung Lee. Path-selection control of a power line inspection robot using sensor fusion [A]. IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems [C],2008:8-13.
[14]干方建,刘正干.基于传感器的操作臂惯性参数在线识别神经网络模型[J].机械科学与技术,2003,22(1)：168-170.
[15]王林松.基于腕力传感器的机器人操作器动力学参数实时识别[D].合肥工业大学,1999.
[16]王冬,张忠林,王臣业.仿生机器蟹单足惯性参数在线识别神经网络模型[J].机械工程师,2006,(4)：36-38.
[17]李祖枢,张华,古建功,陈桂强.3关节单杠体操机器人的动力学参数辨识[J].控制理论与应用,2008,25(2)：242-246.
[18]工北战,何清华,郝鹏等.挖掘机器人动力学参数辨识研究[J].华中师范大学学报(自然科学版),2009,43(2)：239-242.
[19]黄用华PowerCube模块化机器人惯性参数辨识[D].北京工业大学,2006.
[20]陈恩伟.机器人动态特性及动力学参数辨识研究[D].合肥工业大学,2005.
[21]姜春福.基于神经网络的机器人模型辨识与控制研究[D].北京工业大学,2003.
[22]Zheng Li, Yi Ruan. Obstacle negotiation control for a mobile robot suspended on overhead ground wires by optoelectronic sensors [A]. Proceedings of SPIE-The International Society for Optical Engineering [C], Vol.7508,2009.
[23]Li Tang, Lijin Fang, Hongguang Wang. Obstacle-navigation control for a mobile robot suspended on overhead ground wires [A].8th International Conference on Control, Automation, Robotics and Vision [C], Vol.3,2004:2082-2087.
[24]Y. Takahashi, O. Tsubouchi. Tension control of wire suspended mechanism and application to bathroom cleaning robot [A].39th Annual Conference of the Society of Instrument and Control Engineers [C],2000:143-147.
[25]Noritaka Yanai, Motoji Yamamoto, Akira Mohri. Anti-sway control for wire-suspended mechanism based on dynamics compensation [A]. Proceedings of the 2002 IEEE International Conference on Robotics and Automation [C], Vol.4:4287-4292.
[26]董天立,张超平.走钢丝的力学分析.力学与实践,第4期第27卷,2005年：89-90.
[27]邱秉权.分析力学[M].北京：中国铁道出版社,1998.
[28]陈炜,余跃庆,张绪平.欠驱动机器人研究综述[J].机械设计与研究,21(4)2005：22-26.
[29]高丙团,黄学良.欠驱动2DTORA基于部分反馈线性化的非线性控制设计.东南大学学报,第41卷第2期,2011年3月：321-325.
[30]陈玉敏,曲延滨.倒立摆的局部反馈线性化控制方法.微计算机信息,第25卷第8-1期,2009年：4-6.
[31]胡寿松.自动控制原理[M].科学出版社,2007.
[32]李增刚,Adams入门详解与实例.国防工业大学出版社,2006.
[33]TMS320F28335 Data Manual. Texas Instruments Incorporated,2008.
[34]张玉磊,陈进.基于MTi系统的三维运动检测方法的研究[J].测控技术,2008,27(5)：34-36
[35]黄冀等.MTi微惯性航姿系统/GPS组合技术研究[D].哈尔滨工程大学,2009年2月.
[36]ProMotion BDMC2803/3606/3610直流伺服驱动器用户手册.北京博创兴盛机器人技术有限公司,2010.
[37]徐建柱,刁燕,罗华,高山.基于MATLAB和ADAMS的自平衡机器人联合仿真.现代电子技术,第35卷第6期,2012年3月：90-92.
[38]应再恩,平雪良,陈鲁刚.基于ADAMS和MATLAB的双回路PID控制倒立摆联合仿真.机械传动,第8期第36卷,2012年：64-67.