车载线路全断面动态基准测量系统研究
|
摘要
车载线路全断面动态基准测量系统是轨道交通基础设施全断面动态检测系统中一个重点和难点的子系统。掌握基础设施的全断面尺寸及其发展趋势是确保轨道交通安全的基础。车载式全断面动态检测技术由于效率高、对正常运营影响小等优点,受到越来越多的重视。然而在动态检测时,测量设备在运动中由于振动产生六个自由度的不确定性,严重地影响了测量的精度。因此,动态测量设备必须能够自动校准其运动过程中的姿态,即通过动态基准的测量,来修正测量数据。
本文采用惯性导航的原理,通过卡尔曼滤波算法将线路激励模型、车体振动模型、陀螺仪和加速度计的测量模型及数据融合在一起,最大限度地发掘数据之间的相关性,获得高精度的车体运动六自由度姿态信息。将获得的车体姿态信息用以补偿断面测量数据,从而得到准确的全断面尺寸。
论文在确定系统方案的基础上,首先通过详细的推导和计算,建立了车体振动模型;利用三角级数叠加的方法建立了线路的激励模型,结合传感器测量模型得到完整的动态基准测量系统模型。针对线性系统和非线性系统两种情况,分别采用了卡尔曼滤波和无迹卡尔曼滤波的算法在线评估系统状态,得到车体的姿态信息,用以补偿断面测量设备测得的断面尺寸,从而大大提高了动态测量的精度。
Kinematical reference system for vehicular measurement of railway complete profile is an important and difficult subsystem of the scheme for vehicular kinematical measurement system of railway complete profile. Monitoring the status and the trend of railway complete profile is the base of the safe operation of the railway. More and more attention is paid to the technology of vehicular kinematical measurement for its high efficiency and little influence to normal operation. But when the detection is kinematical, there will be six degree of freedom uncertainty caused by the vibration of the measurement equipment, which can seriously affect the measurement accuracy. So the measurement error must be compensated by the kinematical reference system for vehicular measurement to realize the accurate measure of the complete profile.
This paper made full use of the statistic characteristics of the railway line and the vibration characteristics of the vehicle to establish models of the line excitation and the vehicle vibration, and then combined with the measurement model of the sensors to form the integrated system model. Kalman Filter and Unscented Kalman Filter were used to evaluate the system online according to linear system and nonlinear system to get the attitude information of the vehicle. Through this method, we could improve the accuracy of the system. Then, we used the attitude information to compensate the data from the complete profile measurement equipment to get the accurate dimension of the complete profile.
本文采用惯性导航的原理,通过卡尔曼滤波算法将线路激励模型、车体振动模型、陀螺仪和加速度计的测量模型及数据融合在一起,最大限度地发掘数据之间的相关性,获得高精度的车体运动六自由度姿态信息。将获得的车体姿态信息用以补偿断面测量数据,从而得到准确的全断面尺寸。
论文在确定系统方案的基础上,首先通过详细的推导和计算,建立了车体振动模型;利用三角级数叠加的方法建立了线路的激励模型,结合传感器测量模型得到完整的动态基准测量系统模型。针对线性系统和非线性系统两种情况,分别采用了卡尔曼滤波和无迹卡尔曼滤波的算法在线评估系统状态,得到车体的姿态信息,用以补偿断面测量设备测得的断面尺寸,从而大大提高了动态测量的精度。
Kinematical reference system for vehicular measurement of railway complete profile is an important and difficult subsystem of the scheme for vehicular kinematical measurement system of railway complete profile. Monitoring the status and the trend of railway complete profile is the base of the safe operation of the railway. More and more attention is paid to the technology of vehicular kinematical measurement for its high efficiency and little influence to normal operation. But when the detection is kinematical, there will be six degree of freedom uncertainty caused by the vibration of the measurement equipment, which can seriously affect the measurement accuracy. So the measurement error must be compensated by the kinematical reference system for vehicular measurement to realize the accurate measure of the complete profile.
This paper made full use of the statistic characteristics of the railway line and the vibration characteristics of the vehicle to establish models of the line excitation and the vehicle vibration, and then combined with the measurement model of the sensors to form the integrated system model. Kalman Filter and Unscented Kalman Filter were used to evaluate the system online according to linear system and nonlinear system to get the attitude information of the vehicle. Through this method, we could improve the accuracy of the system. Then, we used the attitude information to compensate the data from the complete profile measurement equipment to get the accurate dimension of the complete profile.
引文
铩颷1]蒋兆远等.机车安全运行专家系统.北京.科学出版社.2005
[2]朱力强.隧道限界检测车图像采集系统研制.北京.北方交通大学机械与电气工程学院.1997
[3]周徐昌.沈建森.惯性导航技术的发展及其应用.自动测量与控制.2006.25(9).55-59
[4]黄德鸣.程禄.惯性导航系统.国防工业出版社.1986.1-13
[5]陈哲.捷联惯导系统原理.北京.宇航出版社.1986.143-150
[6]刘志俭.胡小平.贺汉根.GPS定姿系统旋联惯性导航系统组合技术研究.航天控制.2003.1.27-31
[7]Nielson J.T.GPS aided inertial navigation.NAECON.May.1984.406-413
[8]刘建业.袁信.GPS/惯性组合导航开发系统.导航.1994.4.12-18
[9]以光衡.惯性导航原理.北京.航空工业出版社.1987
[10]邓正隆.惯性技术.哈尔滨.哈尔滨工业大学出版社.2006.42
[11]张树侠.孙静.捷联式惯性导航系统.国防工业出版社.1992
[12]J Talaya.R Alamus.E Bosch.Integration of a Terrestrial Laser Scanner with GPS/IMUOrientation Sensors.ISPRS 2004.July.12-23
[13]张成森.铁路机车车辆限界及机车车辆限界的检查方法.铁道技术监督.2004(4).6-7
[14]余祖俊等.轨道车辆瞬时姿态的测量方案研究.电子测量与仪器学报.2007.21(5)
[15]George A.Paulath.吴涛译.惯性级光纤陀螺的研制.惯导与仪表.1997(1).5-9
[16]武凤德.光纤陀螺技术发展综述.舰船导航.2002(4).31-41
[17]许江宁.边少锋.殷立吴.陀螺原理.北京.国防工业出版社.2005
[18]谭显裕.光纤陀螺的发展与应用前景.光纤与电缆及其应用技术.1997(4).49-54
[19]吕宗棋.光纤陀螺.航空电子技术.1995.35-42
[20]刘承香.惯导/GPS/天文组合导航数据融合技术研究.哈尔滨.哈尔滨工程大学.2001
[21]夏禾.张楠.车辆与结构动力相互作用.北京.科学出版社.2005.155-162
[22]刘嘉.车桥耦合振动及其智能控制的研究.武汉.武汉理工大学.2004.11-14
[23]罗林.罗文灿等.干线轨道不平顺紧急补修标准的研究.中国铁道科学.1992(2)
[24]罗文灿.朱开明.罗林.干线轨道不平顺速度管理标准建议值的研究.中国铁道科学.1994(1)
[25]张格明.160km/h线路轨道不平顺管理标准的研究.中国铁道科学.1995(4)
[26]蔡成标.翟婉明.王其昌.轨道几何不平顺安全限值的研究.铁道学报.1995(4)
[27]林玉森.轨道不平顺激励下高速铁路桥上列车走行性研究.西南交通大学.2001.28-46
[28]张有为.维纳与卡尔曼滤波理论导论.人民教育出版社.1980.78-90
[29]崔平远等.基于联合卡尔曼滤波的多传感器信息融合算法及其应用.2001.5(3)
[30]张左民等.卡尔曼滤波设计算法研究进展.控制理论与应用.1995.23.17-27
[31]G.Chen.Approximate Kalman Filtering.World Scientific.1993
[32]张英俊.基于卡尔曼滤波的飞行器定位算法研究.哈尔滨.哈尔滨工业大学.2005.10-11
[33]曲寒冰.卡尔曼滤波器在转台测角系统中的应用.哈尔滨.哈尔滨工业大学.2003.29-31
[34]陈银溢.微惯性航向姿态系统研究.山西.中北大学.2005.57-58
铩颷35]秦永元.卡尔曼滤波与组合导航原理.西安.西北工业大学出版社.1998
[36]Mohinder S G.Application of Kalman filtering to the calibration and alignment of inertialnavigation system.IEEE Translations on Automatic Control.1991.14-18
[37]赵延.机动目标自适应卡尔曼滤波算法研究.成都.电子科技大学.2004.18-20
[38]徐宁寿.随机信号估计与系统控制.北京.北京工业大学出版社.2001.144-146
[39]李涛.非线性滤波方法在导航系统中的应用研究.长沙.国防科技大学.2003
[40]Kushner H J.Dynamical equations for optimum nonlinear filtering.J of Differential Equations.1967.26(3).179-190
[41]Uhlmann J K.Algorithm for multiple target tracking.American Science.1992.80(2).128-141
[42]Sorenson H W.Kalman filtering.Theory and application.New York.IEEE Press.1985
[43]Lerro D.Bar-Shalom Y K.Tracking with Debiased Consistent Converted Measurement vs EKF.IEEE Trans on Aerospace and Electronics Systems.1993.29(3).1015-1022
[44]Maybeck P S.Stochastic models estimation and control.New York.Academic.1982
[45]Caballero-Gil P.Fǘster-Sabater A.A wide family of nonlinear filter functions with a large linearspan.Information Sciences.2003.164(124).197-207
[46]Jazwinski A H.Stochastic processes and filtering theory.New York.Academic.1970
[47]Gordon N J.Salmond D J.Smith AF M.Novel approach to nonlinear/non-Gaussian Bayesianstate estimation.IEEE Proc on Radar and Signal Processing.1993.140(2).107-113
[48]Doucet A.De Freitas A.Gordon N.Sequential monte carlo methods in practice.New York.Springer-Verlag.2001
[49]Arulampalam S.Maskell S.Gordon N.A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking.IEEE Trans on Signal Processing.2002.50(2).174-188
[50]Thrun S.Fox D.Burgard W.Robust monte carlo localization for mobile robots.ArtificialIntelligence.2001.128(122).99-141.
[51]J.Julier.J.K.Uhlmann.A General Method for Approximating Nonlinear Transformations ofProbability Distributions.Technical Report.University of Oxford.1996
[2]朱力强.隧道限界检测车图像采集系统研制.北京.北方交通大学机械与电气工程学院.1997
[3]周徐昌.沈建森.惯性导航技术的发展及其应用.自动测量与控制.2006.25(9).55-59
[4]黄德鸣.程禄.惯性导航系统.国防工业出版社.1986.1-13
[5]陈哲.捷联惯导系统原理.北京.宇航出版社.1986.143-150
[6]刘志俭.胡小平.贺汉根.GPS定姿系统旋联惯性导航系统组合技术研究.航天控制.2003.1.27-31
[7]Nielson J.T.GPS aided inertial navigation.NAECON.May.1984.406-413
[8]刘建业.袁信.GPS/惯性组合导航开发系统.导航.1994.4.12-18
[9]以光衡.惯性导航原理.北京.航空工业出版社.1987
[10]邓正隆.惯性技术.哈尔滨.哈尔滨工业大学出版社.2006.42
[11]张树侠.孙静.捷联式惯性导航系统.国防工业出版社.1992
[12]J Talaya.R Alamus.E Bosch.Integration of a Terrestrial Laser Scanner with GPS/IMUOrientation Sensors.ISPRS 2004.July.12-23
[13]张成森.铁路机车车辆限界及机车车辆限界的检查方法.铁道技术监督.2004(4).6-7
[14]余祖俊等.轨道车辆瞬时姿态的测量方案研究.电子测量与仪器学报.2007.21(5)
[15]George A.Paulath.吴涛译.惯性级光纤陀螺的研制.惯导与仪表.1997(1).5-9
[16]武凤德.光纤陀螺技术发展综述.舰船导航.2002(4).31-41
[17]许江宁.边少锋.殷立吴.陀螺原理.北京.国防工业出版社.2005
[18]谭显裕.光纤陀螺的发展与应用前景.光纤与电缆及其应用技术.1997(4).49-54
[19]吕宗棋.光纤陀螺.航空电子技术.1995.35-42
[20]刘承香.惯导/GPS/天文组合导航数据融合技术研究.哈尔滨.哈尔滨工程大学.2001
[21]夏禾.张楠.车辆与结构动力相互作用.北京.科学出版社.2005.155-162
[22]刘嘉.车桥耦合振动及其智能控制的研究.武汉.武汉理工大学.2004.11-14
[23]罗林.罗文灿等.干线轨道不平顺紧急补修标准的研究.中国铁道科学.1992(2)
[24]罗文灿.朱开明.罗林.干线轨道不平顺速度管理标准建议值的研究.中国铁道科学.1994(1)
[25]张格明.160km/h线路轨道不平顺管理标准的研究.中国铁道科学.1995(4)
[26]蔡成标.翟婉明.王其昌.轨道几何不平顺安全限值的研究.铁道学报.1995(4)
[27]林玉森.轨道不平顺激励下高速铁路桥上列车走行性研究.西南交通大学.2001.28-46
[28]张有为.维纳与卡尔曼滤波理论导论.人民教育出版社.1980.78-90
[29]崔平远等.基于联合卡尔曼滤波的多传感器信息融合算法及其应用.2001.5(3)
[30]张左民等.卡尔曼滤波设计算法研究进展.控制理论与应用.1995.23.17-27
[31]G.Chen.Approximate Kalman Filtering.World Scientific.1993
[32]张英俊.基于卡尔曼滤波的飞行器定位算法研究.哈尔滨.哈尔滨工业大学.2005.10-11
[33]曲寒冰.卡尔曼滤波器在转台测角系统中的应用.哈尔滨.哈尔滨工业大学.2003.29-31
[34]陈银溢.微惯性航向姿态系统研究.山西.中北大学.2005.57-58
铩颷35]秦永元.卡尔曼滤波与组合导航原理.西安.西北工业大学出版社.1998
[36]Mohinder S G.Application of Kalman filtering to the calibration and alignment of inertialnavigation system.IEEE Translations on Automatic Control.1991.14-18
[37]赵延.机动目标自适应卡尔曼滤波算法研究.成都.电子科技大学.2004.18-20
[38]徐宁寿.随机信号估计与系统控制.北京.北京工业大学出版社.2001.144-146
[39]李涛.非线性滤波方法在导航系统中的应用研究.长沙.国防科技大学.2003
[40]Kushner H J.Dynamical equations for optimum nonlinear filtering.J of Differential Equations.1967.26(3).179-190
[41]Uhlmann J K.Algorithm for multiple target tracking.American Science.1992.80(2).128-141
[42]Sorenson H W.Kalman filtering.Theory and application.New York.IEEE Press.1985
[43]Lerro D.Bar-Shalom Y K.Tracking with Debiased Consistent Converted Measurement vs EKF.IEEE Trans on Aerospace and Electronics Systems.1993.29(3).1015-1022
[44]Maybeck P S.Stochastic models estimation and control.New York.Academic.1982
[45]Caballero-Gil P.Fǘster-Sabater A.A wide family of nonlinear filter functions with a large linearspan.Information Sciences.2003.164(124).197-207
[46]Jazwinski A H.Stochastic processes and filtering theory.New York.Academic.1970
[47]Gordon N J.Salmond D J.Smith AF M.Novel approach to nonlinear/non-Gaussian Bayesianstate estimation.IEEE Proc on Radar and Signal Processing.1993.140(2).107-113
[48]Doucet A.De Freitas A.Gordon N.Sequential monte carlo methods in practice.New York.Springer-Verlag.2001
[49]Arulampalam S.Maskell S.Gordon N.A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking.IEEE Trans on Signal Processing.2002.50(2).174-188
[50]Thrun S.Fox D.Burgard W.Robust monte carlo localization for mobile robots.ArtificialIntelligence.2001.128(122).99-141.
[51]J.Julier.J.K.Uhlmann.A General Method for Approximating Nonlinear Transformations ofProbability Distributions.Technical Report.University of Oxford.1996