机动目标跟踪关键技术研究
|
摘要
随着科学技术的迅速发展和现代战争战场环境日趋复杂，现代目标的机动性能越来越复杂多变，而人们对跟踪的性能要求则越来越高，因而，使得现代机动目标跟踪技术成为无论是在军事上还是民用上的都是一个研究热点。在目标跟踪技术中，目标的状态最优估计算法是跟踪技术有效实现的数学工具和保证，目标的运动建模是系统的基础和核心，而系统的结构则是整个跟踪算法的有效实现手段，因此，本文从上述三个方面对机动目标跟踪技术进行了研究，论文的主要工作有：
分析了现代目标跟踪系统的非线性特性，对目前常用的几种非线性滤波方法扩展卡尔曼滤波算法、无迹卡尔曼滤波算法和容积卡尔曼滤波算法进行理论推导，分析并仿真验证了这三种非线性滤波算法在高维环境下对非线性状态的估计精度性能和稳定性能为后续非线性机动目标跟踪系统的自适应算法研究提供了相应的理论依据。
阐述了当前统计模型及其自适应的目标跟踪算法理论和性能，指出当前统计模型中反映目标机动性能的机动频率初始值难以确定、且其不随目标机动性能变化而自适应调整从而造成当前统计模型对机动性能变化的目标描述能力较差，降低目标跟踪系统的整体跟踪精度，提出了一种基于机动性能检测的机动频率自适应算法，建立机动频率自适应调节模型，分析并仿真验证了在目标加速度反复机动变化情况下，该算法对目标机动频率的自适应调节能力，并与传统固定机动频率参数的当前统计模型算法进行了比较分析。
指出当前统计模型采用修正的瑞利分布描述目标的加速度存在一定的局限，使得当前统计模型对弱机动和非机动时算法跟踪性能较差，提出了一种加速度极值自适应滤波算法，采用一种铃形模糊隶属函数修正加速度极值，同时为了增强改进算法对加速度突变情况的跟踪能力，采用时变渐消因子调节修正了的加速度极值。分析并仿真验证了当目标在机动过程中存在弱机动和非机动运动，同时也存在加速度突变机动情况时该改进算法对目标的跟踪效果，并与传统当前统计模型和文献改进算法进行了比较分析。
阐述了基于Jerk模型的目标跟踪理论及其跟踪性能，指出Jerk模型在目标的加速度变化率阶跃变化时系统会产生均值不为零的稳态误差，大大降低了Jerk模型的跟踪性能，提出了一种基于MEP-Jerk模型机动目标跟踪自适应算法，采用模型预测滤波预测估计Jerk模型的模型误差进行，实时修正Jerk模型对系统噪声假设不全面所造成的影响，同时针对非线性目标跟踪系统，采用容积卡尔曼滤波器实现上述自适应算法的状态估计。分析并仿真验证了当目标在运动过程中存在非机动运动、常规转弯运动以及加速度变化率阶跃机动时，该自适应算法和容积卡尔曼滤波器对目标跟踪系统的影响。
指出了Jerk模型中机动频率和机动方差需要事前设置且不能自适应调整，降低了Jerk模型对机动目标跟踪精度和鲁棒性能，提出了一种基于AR模型的Jerk参数自适应估计算法，通过建立MS模型，并根据AR模型参数估计方法，实时的估计目标的Jerk机动频率和机动方差值。同时针对非线性滤波算法中容积卡尔曼滤波器在滤波过程具有较大计算量甚至可能出现不稳定的情况，提出采用平方根容积卡尔曼滤波器完成上述参数自适应算法的各状态滤波。分析并仿真验证了当目标反复做Jerk机动运动时该算法对目标跟踪系统的影响，并与基于扩展卡尔曼滤波的Jerk模型算法以及基于SRCKF的Jerk模型算法进行了比较分析。
阐述了多模型理论和交互式多模型理论，详细分析了变结构交互式多模型算法中的可能模型集自适应算法和期望模型集自适应算法，指出可能模型集和期望模型集自适应算法都不能解决系统模型集无法覆盖目标的所有运动状态这一问题，提出了一种基于Kullback-Leiber信息因子分析多模型算法中模型集各模型之间匹配程度的算法，并在此基础上实现了系统模型集进行自适应调节。同时，针对变结构多模型算法中模型集中各模型在各时刻其模型概率和模型之间的转换矩阵都会发生变化，提出了基于Kullback-Leiber信息因子的模型概率更新算法，推导了更新后模型转换概率矩阵中各元素值的计算方程。分析并仿真验证了当目标反复进行参数变化的机动运动时该算法对机动目标跟踪的能力，并与传统的固定结构多模型算法进行了比较分析。
With the rapid development of the technology and science, the environment of themodern war has been more and more complex. The maneuverability of targets is becomingmore and more complicated and inconstant; meanwhile the demand of the trackingperformance index also is increasing. Hence, the targets tracking technology becomes theresearch hinge in the military and civil. In the targets tracking technology, the state optimalestimation algorithm is the mathematical tool and guarantee for realizing the trackingtechnology. The modeling of the target is the foundation and the kernel of the trackingsystems. The system configuration design is the effective mean to implement the trackingalgorithm. Therefore, this dissertation studies for the maneuvering target tracking from thethree statements above, which can be summarized as follows:
The nonlinear character of the modern targets tracking system has been analyzed. Thenonlinear filter methods that include Extend Kalman Filter (EKF), Unscented Kalman Filter(UKF) and Cubature Kalman Filter (CKF) are elaborated in theory. Then the precision andthe stabilization performance of state estimation in the high dimensional situation wasanalyzed and verified by simulation, which provides the theoretical basis for the adaptivealgorithm research on the nonlinear maneuver targets tracking system subsequently.
Then theoretic analyzing and performance of the current statistical model and itsadaptive maneuver targets tracking was elaborated and the problem that the maneuveringfrequency of the targets maneuvering was hardly defined accurately and the value can adjustwith the change of the targets maneuvering was point out, which caused the currentstatistical model can’t describe the target effectively and decrease the precision of targettracking systems. Then a maneuvering frequency adaptive algorithm based on maneuveringdetection is presented. The model of maneuvering frequency adaptive adjusted wasconstructed. Then the algorithm performance was analyzing and verified by simulation withtradition algorithm as the targets acceleration change repeatedly.
The description of the acceleration of the target in the current statistical model wasmodified Raleigh distribution, which made the model has poor performance on weak andnon-motorized maneuvering targets. Then an adaptive filter algorithm based onextremum adaptive of acceleration was proposed, which a bell shape function as fuzzymembership function was utilized to adjust the upper and lower limits of target acceleration.At the same time, to enhance the response capability of the model as sudden maneuver or the acceleration changed greatly, a fading factor was proposed to adjust revised extremevalue of acceleration. The analysis and simulation results show that the algorithm has abetter performance on tracking weak and non-maneuvering maneuvering targets than thetraditional algorithm.
The theory of target tracking and its performance based on Jerk model wasexpatiated and then that the system will have non-zero steady state error as the input signalof the target acceleration rate changed by jump function. That greatly reduces the trackingperformance of the Jerk model. An adaptive algorithm based on the MEP-on Jerk model formaneuvering target tracking then was proposed.By using the model prediction filter topredict the estimated model error of Jerk model, the influence caused by not fully estimatethe Jerk model is corrected online. At the same time the CKF was used to estimate the statefor nonlinear target tracking system. Analysis and simulation show that the algorithmperformance and the traditional algorithm as the target running with the non-motorizedsports, conventional turning movement and acceleration of the rate of change of the stepmaneuvers.
Then the paper pointed out that the frequency and variance of maneuvering need topre-define and the value won’t be changed when the target maneuvering that makes theeffect of tracking drop down dramatically. A Jerk parameters adaptive estimation algorithmbased on the AR model was proposed. According to the parameter estimation method of theAR model with the MS model, the Jerk maneuvering frequency and variance of targetestimates real-time. At the same time as the CKF has complex account and sometime evenmaking the system divergence, the SRCKF was utilized. The analysis and simulation showthat the performance of this algorithm when the target running in the Jerk motion, comparedwith the Jerk model based on EKF algorithm and SRCKF algorithm.
The general multiple model theory and the interacted multiple model theory wereintroduced. Then the Likely-Model Set algorithm (LMS) and the Expected-ModelAugmentation (EMA) of variable interacted multiple model was illuminate particularly. Asthe LMS and EMA all can’t cover all the real motion of target, a algorithm based onKullback-Liber (K-L) to analyze the model match degree with the others. Based on this, themodel set adaptive adjustment was implemented. As the model probability and thetransition matrix of model sets was changed in variable multiple model algorithm, a modelprobability updated algorithm was proposed and the equations to compute elements of thetransition matrix was deduced. The analysis and simulation show that the performance ofthis algorithm as the parameters changed all the time in target tracking.
分析了现代目标跟踪系统的非线性特性，对目前常用的几种非线性滤波方法扩展卡尔曼滤波算法、无迹卡尔曼滤波算法和容积卡尔曼滤波算法进行理论推导，分析并仿真验证了这三种非线性滤波算法在高维环境下对非线性状态的估计精度性能和稳定性能为后续非线性机动目标跟踪系统的自适应算法研究提供了相应的理论依据。
阐述了当前统计模型及其自适应的目标跟踪算法理论和性能，指出当前统计模型中反映目标机动性能的机动频率初始值难以确定、且其不随目标机动性能变化而自适应调整从而造成当前统计模型对机动性能变化的目标描述能力较差，降低目标跟踪系统的整体跟踪精度，提出了一种基于机动性能检测的机动频率自适应算法，建立机动频率自适应调节模型，分析并仿真验证了在目标加速度反复机动变化情况下，该算法对目标机动频率的自适应调节能力，并与传统固定机动频率参数的当前统计模型算法进行了比较分析。
指出当前统计模型采用修正的瑞利分布描述目标的加速度存在一定的局限，使得当前统计模型对弱机动和非机动时算法跟踪性能较差，提出了一种加速度极值自适应滤波算法，采用一种铃形模糊隶属函数修正加速度极值，同时为了增强改进算法对加速度突变情况的跟踪能力，采用时变渐消因子调节修正了的加速度极值。分析并仿真验证了当目标在机动过程中存在弱机动和非机动运动，同时也存在加速度突变机动情况时该改进算法对目标的跟踪效果，并与传统当前统计模型和文献改进算法进行了比较分析。
阐述了基于Jerk模型的目标跟踪理论及其跟踪性能，指出Jerk模型在目标的加速度变化率阶跃变化时系统会产生均值不为零的稳态误差，大大降低了Jerk模型的跟踪性能，提出了一种基于MEP-Jerk模型机动目标跟踪自适应算法，采用模型预测滤波预测估计Jerk模型的模型误差进行，实时修正Jerk模型对系统噪声假设不全面所造成的影响，同时针对非线性目标跟踪系统，采用容积卡尔曼滤波器实现上述自适应算法的状态估计。分析并仿真验证了当目标在运动过程中存在非机动运动、常规转弯运动以及加速度变化率阶跃机动时，该自适应算法和容积卡尔曼滤波器对目标跟踪系统的影响。
指出了Jerk模型中机动频率和机动方差需要事前设置且不能自适应调整，降低了Jerk模型对机动目标跟踪精度和鲁棒性能，提出了一种基于AR模型的Jerk参数自适应估计算法，通过建立MS模型，并根据AR模型参数估计方法，实时的估计目标的Jerk机动频率和机动方差值。同时针对非线性滤波算法中容积卡尔曼滤波器在滤波过程具有较大计算量甚至可能出现不稳定的情况，提出采用平方根容积卡尔曼滤波器完成上述参数自适应算法的各状态滤波。分析并仿真验证了当目标反复做Jerk机动运动时该算法对目标跟踪系统的影响，并与基于扩展卡尔曼滤波的Jerk模型算法以及基于SRCKF的Jerk模型算法进行了比较分析。
阐述了多模型理论和交互式多模型理论，详细分析了变结构交互式多模型算法中的可能模型集自适应算法和期望模型集自适应算法，指出可能模型集和期望模型集自适应算法都不能解决系统模型集无法覆盖目标的所有运动状态这一问题，提出了一种基于Kullback-Leiber信息因子分析多模型算法中模型集各模型之间匹配程度的算法，并在此基础上实现了系统模型集进行自适应调节。同时，针对变结构多模型算法中模型集中各模型在各时刻其模型概率和模型之间的转换矩阵都会发生变化，提出了基于Kullback-Leiber信息因子的模型概率更新算法，推导了更新后模型转换概率矩阵中各元素值的计算方程。分析并仿真验证了当目标反复进行参数变化的机动运动时该算法对机动目标跟踪的能力，并与传统的固定结构多模型算法进行了比较分析。
With the rapid development of the technology and science, the environment of themodern war has been more and more complex. The maneuverability of targets is becomingmore and more complicated and inconstant; meanwhile the demand of the trackingperformance index also is increasing. Hence, the targets tracking technology becomes theresearch hinge in the military and civil. In the targets tracking technology, the state optimalestimation algorithm is the mathematical tool and guarantee for realizing the trackingtechnology. The modeling of the target is the foundation and the kernel of the trackingsystems. The system configuration design is the effective mean to implement the trackingalgorithm. Therefore, this dissertation studies for the maneuvering target tracking from thethree statements above, which can be summarized as follows:
The nonlinear character of the modern targets tracking system has been analyzed. Thenonlinear filter methods that include Extend Kalman Filter (EKF), Unscented Kalman Filter(UKF) and Cubature Kalman Filter (CKF) are elaborated in theory. Then the precision andthe stabilization performance of state estimation in the high dimensional situation wasanalyzed and verified by simulation, which provides the theoretical basis for the adaptivealgorithm research on the nonlinear maneuver targets tracking system subsequently.
Then theoretic analyzing and performance of the current statistical model and itsadaptive maneuver targets tracking was elaborated and the problem that the maneuveringfrequency of the targets maneuvering was hardly defined accurately and the value can adjustwith the change of the targets maneuvering was point out, which caused the currentstatistical model can’t describe the target effectively and decrease the precision of targettracking systems. Then a maneuvering frequency adaptive algorithm based on maneuveringdetection is presented. The model of maneuvering frequency adaptive adjusted wasconstructed. Then the algorithm performance was analyzing and verified by simulation withtradition algorithm as the targets acceleration change repeatedly.
The description of the acceleration of the target in the current statistical model wasmodified Raleigh distribution, which made the model has poor performance on weak andnon-motorized maneuvering targets. Then an adaptive filter algorithm based onextremum adaptive of acceleration was proposed, which a bell shape function as fuzzymembership function was utilized to adjust the upper and lower limits of target acceleration.At the same time, to enhance the response capability of the model as sudden maneuver or the acceleration changed greatly, a fading factor was proposed to adjust revised extremevalue of acceleration. The analysis and simulation results show that the algorithm has abetter performance on tracking weak and non-maneuvering maneuvering targets than thetraditional algorithm.
The theory of target tracking and its performance based on Jerk model wasexpatiated and then that the system will have non-zero steady state error as the input signalof the target acceleration rate changed by jump function. That greatly reduces the trackingperformance of the Jerk model. An adaptive algorithm based on the MEP-on Jerk model formaneuvering target tracking then was proposed.By using the model prediction filter topredict the estimated model error of Jerk model, the influence caused by not fully estimatethe Jerk model is corrected online. At the same time the CKF was used to estimate the statefor nonlinear target tracking system. Analysis and simulation show that the algorithmperformance and the traditional algorithm as the target running with the non-motorizedsports, conventional turning movement and acceleration of the rate of change of the stepmaneuvers.
Then the paper pointed out that the frequency and variance of maneuvering need topre-define and the value won’t be changed when the target maneuvering that makes theeffect of tracking drop down dramatically. A Jerk parameters adaptive estimation algorithmbased on the AR model was proposed. According to the parameter estimation method of theAR model with the MS model, the Jerk maneuvering frequency and variance of targetestimates real-time. At the same time as the CKF has complex account and sometime evenmaking the system divergence, the SRCKF was utilized. The analysis and simulation showthat the performance of this algorithm when the target running in the Jerk motion, comparedwith the Jerk model based on EKF algorithm and SRCKF algorithm.
The general multiple model theory and the interacted multiple model theory wereintroduced. Then the Likely-Model Set algorithm (LMS) and the Expected-ModelAugmentation (EMA) of variable interacted multiple model was illuminate particularly. Asthe LMS and EMA all can’t cover all the real motion of target, a algorithm based onKullback-Liber (K-L) to analyze the model match degree with the others. Based on this, themodel set adaptive adjustment was implemented. As the model probability and thetransition matrix of model sets was changed in variable multiple model algorithm, a modelprobability updated algorithm was proposed and the equations to compute elements of thetransition matrix was deduced. The analysis and simulation show that the performance ofthis algorithm as the parameters changed all the time in target tracking.
引文
[1] X. Rong Li, Vesselin P. Survey of Maneuvering Target Tracking. Part I: DynamicModels [J]. IEEE Transactions on Aerospace and Electronic Systems,2003,39(4):1333-1362P.
[2] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part II: BallisticTargets Models [C].//Proceeding of SPIE Conference on Singal and Data Processingof Small Targets.San Diego, CA, USA,2001(4473-63):1-23P.
[3] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part III:Meauserment Models [C].//Proceeding of SPIE Conference on Singal and DataProcessing of Small Targets.San Diego, CA, USA,2001(4473-41):1-24P.
[4] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part IV:Decision-Based Methods [C].//Proceeding of SPIE Conference on Singal and DataProcessing of Small Targets.Orlando, FL, USA,2002(4428-60):1-24P.
[5] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part V:Multiple-Model Methods [J]. Submited to IEEE Transaction on Aerospace andElectronic Systems,2003:1-58P.
[6]权太范.目标跟踪新理论与技术[M].北京:国防工业出版社,2009.
[7]潘泉,梁彦,杨峰等.现代目标跟踪与信息融合[M].北京:国防工业出版社,2009.
[8]何友,修建娟,张晶炜等.雷达数据处理及应用(第二版)[M].北京:电子工业出版社,2009.
[9]杨元喜.自适应动态导航定位[M].北京:测绘出版社,2006.
[10]杨峰.现代目标跟踪与多传感器融合关键技术研究[D].西北工业大学工学博士学位论文,2006
[11]夏佩伦编著,目标跟踪与信息融合[M].北京:国防工业出版社,2010.
[12]梁彦,潘泉,杨峰等.复杂系统的现代估计理论及应用[M].北京:科学出版社,2009.
[13]何友,王国宏,陆大紟等.多传感器信息融合及应用(第二版)[M].北京:电子工业出版社,2007.
[14]周宏仁,敬忠良,王培德.机动目标跟踪[M].北京:国防工业出版社,1991.
[15]彭东亮,文成林,薛安克.多传感器多源信息融合理论及应用[M].北京:科学出版社,2010.
[16]石章松,刘忠.目标跟踪与数据融合理论及方法[M].北京：国防工业出版社,2010.
[17]史玉柱.最优估计的计算方法[M].北京:科学出版社,2001.
[18] Jazwinski A H. Stochastic processes and filtering theory [M]. New York: Academic,1970.
[19] Maybeck P S. Stochastic Models，Estimation and Control [M]. New York: AcademicPress,1979.
[20] Sorenson H W. Kalman filtering: Theory and Application [M]. New York: IEEE Press,1985.
[21]俞济祥.卡尔曼滤波及其在惯性导航中的应用[M].北京:航空专业教材编审组,1984.
[22]秦永元,张洪钺,汪叔华.卡尔曼滤波与组合导航原理[M].西安：西北工业大学出版社,1998.
[23]付梦印,邓志红,闫莉萍. Kalman滤波理论及其在导航系统中的应用(第二版)[M].北京：科学出版社,2010.
[24] Dan Simon. Optimal State Estimation [M]. John Wiley&Sons, Inc., Hoboken, NewJersey,2006.
[25] Julier S J, Uhlmann J K. A new approach for filtering nonlinear system[C].//Proceedings of the1995American Control Conference,1995:1628-1632P.
[26] Julier S J, Uhlmann J K. A new method for the nonlinear transformation of means andcovariances in filters and estimators [J]. IEEE Transactions on Automatic Control,2000,45(3):477-482P.
[27] N rgarrd M, Poulsen N K, Ravn O. New developments in state estimation fornonlinear systems [J]. Automatica,2000,36(11):1627-1638P.
[28] Merwe R V D. Sigma-Point Kalman Filters for Probabilistic Inference in DynamicState-Space Models [J]. http://www.cslu.ogi.edu/publications,2004.
[29] Julier S J, Uhlmann J K. A general method for approximating nonlineartransformation of probability distributions [J]. http://www.eng.ox.ac.uk,1996.
[30] Julier S J, Uhlmann J K. Reduced sigma point filters for the propagation of means andcovariances through nonlinear transformations [C].//Proceedings of AmericanControl Conf. Jefferson City,2002:887-892P.
[31] Merwe R V D, Wan E A. The Square-root Unscented Kalman Filter for State andParameter-eatimation [C].//International Conference on Acoustics, Speech, andSignal Processing. Salt Lake City,2001:3461-3464P.
[32] I. Arasaratnam, S. Haykin. Cubature Kalman filters [J]. IEEE Trans AutomaticControl.2009,54:1254-1269P.
[33] Ienkaran Arasaratnam, Simon Haykin. Cubature Kalman Filtering forContinuous-Discrete Systems: Theory and Simulations [J]. IEEE Transactions onSignal Processing,2010,58:4977-4993P.
[34] Ienkaran Arasaratnam, Simon Haykin. Cubature Kalman smoothers [J].Automatica(2011),doi:10.1016/j.automatica.2011.8.005
[35]兰剑,慕春棣.基于参考加速度的机动目标跟踪模型[J].清华大学学报,2008,48(10):1549-1552页.
[36]肖雷,刘高峰,魏建仁.几种机动目标运动模型的跟踪性能对比[J].火力与指挥控制,2007,(5):106-109页.
[37] Su Y P. The Research on the Maneuvering Object base on Adaptive followingarithmetic[C].//2010International Conference on Intelligent Computation Technologyand Automation,2010:938-941P.
[38] R.L.T.Hampton, J.R.Cooke. Unsupervised tracking of maneuvering vehicles [J], IEEETransaction on Aerospace and Electronic Systems,1973,9(2):197-297P.
[39] Friedland B. Optimum steady-state position and velovity estimation using noisysampled position data [J]. IEEE Transaction on Aerospace and Electronic Systems,1973,9(6):906-111P.
[40] Signger R A. Estimating optimal tracking filter performance for manned maneuveringtargets [J]. IEEE Trans. on Aerospace and Electromic Systems,1970,6(7):473-183P.
[41] R.L.Moose. An adaptive state estimation solution to the maneuvering target problem[J]. IEEE Transactions on Automatic Control,1975,20(6):359-362P.
[42] Zhuo H, Kumar K S P. A “current” statistical model and adaptive algorithm forestimating maneuvering targets [J]. AIAA Journal of Guidance,1984,7(5):596-602P.
[43]张伯彦,蔡庆宇,袁曾任.一种跟踪机动目标的新方法[J].系统工程与电子技术，1990,12(1):45-55页.
[44] Mehrotra K, Mahapatra P R. A Jerk Model for Tracking Highly Maneuvering Targets[J]. IEEE Trans. on Aerospace and Electronic Systems,1997,33(4):1094-1104P
[45] S. Koteswara Rao. Comments on “A Jerk model for highly maneuvering Targets”[J].IEEE transaction on aerospace and electronic system,1998,34(3):982-983P.
[46] Pravas R.Mahapatra. Kishore Mehrotra Mix coordinates tracking of generalizedmaneuvering targets using acceleration and Jerk models [J]. IEEE transaction onaerospace and electronic system,2000,36(3):992-1000P.
[47] GA.Waston, W.D.Blair. IMM algorithm for tracking targets that maneuvering throughcoordinated turns[C].//Proceedings of SPIE Signal and Data Processing of SmallTargets,1992:236-247P.
[48] R.A.Best, J.P.Norton. A new model and efficient tracker for a target with curvilinearmotion [J]. IEEE transaction on aerospace and electronic system,1997,33(31):1030-1037P.
[49] T.Matsuzaki, H.Kameda, S.Tsujimichi.ect. Maneuvering target tracking using constantvelocity and constant angular velocity model[C].//SICE’99,1999:1135-1138P.
[50] J.P.Herferty. Maneuvering Targets: The Use of Turn-rate Distributions for AccretionModeling [J].IEEE Trans. on Aerospace and Electronic Systems,1996,32(4):1355-1361P.
[51] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part II: BallisticTargets Models[C].//Proceeding of SPIE Conference on Singal and Data Processingof Small Targets.San Diego, CA, USA,2001(4473-63):1-23P.
[52]范小军,刘峰.一种新的机动目标跟踪的多模型算法[J].电子与信息学报,2007,29(3):532-535页.
[53]曹梦龙,崔平远.多模型高精度组合导航算法研究[J].系统工程与电子技术，2008,30(7):1304-1307页.
[54]梁彦,程咏梅,贾宇岗.交互式多模型算法性能分析[J].控制理论与应用,2001,18(4):487-492页.
[55]黄翔宇,彭冬亮.一种变结构多模型机动目标跟踪算法的研究[J].杭州电子科技大学学报,2010,30(4):195-199页.
[56]李俊,周凤歧,周军.机动目标跟踪的多模型多机动策略算法[J].系统仿真学报，2009,21,(3):668-671页.
[57]马海平,陈子栋.一种基于H∞滤波的模糊变结构交互式多模型算法[J].电子学报，2008,36(2):245-249页.
[58]徐肖豪,高彦杰,杨国庆.交互式多模型算法中模型集选择的分析研究[J].航空学报,2004,25(4):352-356页.
[59] X. Rong Li, Vesselin P. Jilkov. Multiple-Model Estimation with Variable Structure [J].IEEE Transaction on Automatic Control,1996,41(4):478-493P.
[60] X. Rong Li. Multiple-Model Estimation with Variable Structure II: Model-SetAdaptation [J]. IEEE Transaction on Automatic Control,2000,45(11):2047-2060P.
[61] X. Rong Li, Xiaorong Zhi, Youmin Zhang. Multiple-Model Estimation with VariableStructure III: Model-Group Switching Algorithm [J]. IEEE Transactions on Aerospaceand Electronic Systems,1999,35(1):225-241P.
[62] X. Rong Li, Youmin Zhang, Xiaorong Zhi. Multiple-Model Estimation with VariableStructure IV: Design and Evaluation of Model-Group Switching Algorithm [J]. IEEETransactions on Aerospace and Electronic Systems,1999,35(1):242-254P.
[63] X. Rong Li, Youmin Zhang. Multiple-Model Estimation with Variable Structure V:Likely-Model Set Algorithm [J]. IEEE Transactions on Aerospace and ElectronicSystems,2000,36(2):448-466P.
[64] X. Rong Li, Vesselin P. Jilkov, Jifeng Ru. Multiple-Model Estimation with VariableStructure VI: Expected-Mode Augmentation [J]. IEEE Transactions on Aerospace andElectronic Systems,2005,41(3):853-867P.
[65] X.Rong Li, Vesselin P.Jilkov. Survey of Maneuvering Target Tracking Part II: MotionModels of Ballistic and Space Targets [J]. IEEE Transactions on Aerospace andElectronic Systems,2010,46(1):96-119P.
[66] Jaipal R. Katkuri, Vesselin P. Jilkov, X. Rong Li. A Comparative Study of NonlinearFilters for Target Tracking in Mixed Coordinates [C].//42nd South EasternSymposium on System Theory Tyler, TX, USA,2010:202-207P.
[67] Huiming Chen, X. Rong Li, Yaakov Bar-Shalom. On Joint Track Initiation andParameter Estimation under Measurement Origin Uncertainty [J]. IEEE Transactionson Aerospace and Electronic Systems,2004,4(20):675-694P.
[68]杨柏胜.被动多传感器探测目标跟踪技术研究[D].西安电子科技大学工学博士学位论文,2009.
[69]彭亮.机动目标跟踪算法研究[D].西北工业大学工学博士学位论文,2007.
[70]张高煜.目标跟踪与多传感器信息融合若干问题研究[D].西安电子科技大学工学博士学位论文,2006.
[71]乔向东.信息融合系统中目标跟踪技术[D].西安电子科技大学工学博士学位论文,2003.
[72]孙福明.机动目标跟踪状态估计与数据关联技术研究[D].中国科学技术大学工学博士学位论文,2007.
[73]罗笑冰.强机动目标跟踪技术研究[D].国防科学技术大学工学博士学位论文,2007.
[74] Smith A F M, Geland A E. Bayesian Statistics without tears: a sampling-resamplingperspective [J]. The American Statistican.1992,46(2):84-88P.
[75] Haykin S, Huber K, Chen Z. Bayesian sequential state estimation for MIMO wirelesscommunications [C].//Proceedings of the IEEE,2004,92(3):439-454P.
[76] Handschin J E. Monte Carlo techniques for prediction and filtering of nonlinearstochastic processes [J]. Automatica,1970,6(3):555-563P.
[77]王小旭.非线性SPKF滤波算法研究及其在组合导航中的应用[D].哈尔滨工程大学工学博士学位论文,2010.
[78]郝钢.自校正观测融合Kalman估值器及其在典型跟踪系统中的应用[D].哈尔滨工程大学工学博士学位论文,2011.
[79]李涛.非线性滤波方法在导航系统中的应用研究[D].国防科技大学工学博士学位论文,2003.
[80]杨萌.非线性滤波及其在惯性导航系统传递对准中应用研究[D].哈尔滨工程大学工学博士学位论文,2010.
[81]聂琦.非线性滤波及其在导航系统中的应用[D].哈尔滨工程大学工学博士学位论文,2008.
[82]尹建军.线性/非线性系统的混合动态滤波理论及应用[D].复旦大学工学博士学位论文,2008.
[83] Weisheng Wu, Chunlei Song, Junhou Wang. Cubature Gaussian Particle Filter ForInitial Alignment of Strapdown Inertial Navigation System [C].//2010FirstInternational Conference on Pervasive Computing, Signal Processing andApplications,2010:1196-1200P.
[84] H. Pesonen, R. Pich′e. Cubature-based Kalman Filters for Positioning [C].//20107thWorkshop on Positioning Navigation and Communication (WPNC),2010:45-49P.
[85] Davide Macagnano, Giuseppe Thadeu Freitas de Abreu. Multitarget Tracking with theCubature Kalman Probability Hypothesis Density Filter [J].2010Conference Recordof the Forty Fourth Asilomar Conference on Signals, Systems and Computers(ASILOMAR),2010:1455-1459P.
[86] Kumar Pakki. Bharani Chandra, Da-Wei Gu.ect. Cubature Information Filter and itsApplications[C].//2011American Control Conferenceon O'Farrell Street, SanFrancisco, CA, USA,2011:3609-3614P.
[87] Jing Mu, Yuan-li Cai. Iterated Cubature Kalman Filter and its Application[C].//Proceedings of the2011IEEE International Conference on Cyber Technology inAutomation, Control, and Intelligent Systems, Kunming, China,2011:33-37P.
[88] Chen Hai, Shan Ganlin. Attitude Angle Aided IMMCKF Algorithm [C].//The TenthInternational Conference on Electronic Measurement&Instruments,2011:197-200P.
[89]郝利超,雷虎民,王飞等.基于当前统计模型的模糊自适应跟踪算法[J].电光与控制,2009,16(10):18-21页.
[90]任少伟,王睿,张平定.基于机动频率自适应的目标跟踪算法[J].空军工程大学学报(自然科学版),2004,5(5):32-35页.
[91]王军政,沈伟,赵红波.机动目标跟踪中机动频率的自适应调整[J].北京理工大学学报,2007,27(1):38-41页.
[92]陈佳俊,刘高峰,辛晋生等.机动频率模糊自适应目标跟踪算法研究[J].弹箭与制导学报,2010,30(2):259-262页.
[93]刘高峰,顾雪峰,王华楠.跟踪反舰导弹的机动频率自适应算法[J].计算机仿真，2008,25(10):59-61页.
[94]刘昌云,刘进忙,陈长兴等.机动目标跟踪的机动频率自适应算法[J].控制理论与应用,2004,21(6):961-965页.
[95] WANG W, HOU H L. An improved Current Statistical Model for maneuvering targettracking[C].//4th IEEE Conference on Industrial Electronics and Applications,2009:4017-4020P.
[96] Wang Wei， Huo Hong-lu. An improved Current Statistical Model for maneuveringTargets Tracking[C].//4th IEEE Conference on Industrial Electronics andApplications,2009:4017-4020P.
[97]王炜,杨露菁.一种改进的当前统计模型及其自适应滤波算法[J].系统仿真学报,2005,17(8):2037-2040页.
[98]李辉,沈莹,张安等.机动目标跟踪中一种新的自适应滤波算法[J].西北工业大学学报,2006,24(3):354-357页.
[99]段战胜,韩崇昭,党宏社.最大加速度未知的当前统计模型机动目标跟踪[J].计算机工程与应用,2003,(29):19-22页.
[100]巴宏欣,赵宗贵,杨飞等.机动目标的模糊自适应跟踪算法[J].系统仿真学报,2004,16(6):1181-1186页.
[101]王炜,杨露菁.一种改进的当前统计模型及其自适应滤波算法[J].系统仿真学报,2005,(3):2037-2040页.
[102]胡洪涛,敬忠良,田宏伟等.基于“当前”统计模型的模糊自适应跟踪算法[J].系统仿真学报,2005,17(2):293-299页.
[103]王向华,覃征,杨慧杰等.基于“当前”统计模型的模糊自适应跟踪算法[J].兵工学报,2009,(8):1089-1093页
[104]刁联旺,杨静宇.一种改进的机动目标“当前”统计模型的描述[J].兵工学报,2005,26(6):625-628页.
[105]胡振涛,刘先省.基于当前统计模型的一种改进机动目标跟踪算法[J].山东大学学报(工学版),2005,35(3):111-114页.
[106] Xiaoxiao Wu，Fuchun Sun，Wansen Wang.ect. Adaptive algorithm of trackingmaneuvering targets using rate of acceleration [C].//Proceeding of the6th WorldCongress on Intelligent Control and Automation Dalian,2006:1838-1841P.
[107] Yu H X, Fu C K, Jiang L. A fuzzy adaptive tracking algorithm based on currentstatistical probabilistic data association[C].//2nd International Conference on SignalProcessing Systems,2010:757-759P.
[108] Wang L H, Zhu Q D, Xing Z Y. Adaptive Nonlinear Filter Algorithm Based OnCurrent Statistical Model[C].//Proc. of the2007IEEE International conference onMechatronics and automation, Harbin,2007:2414-2418P.
[109] Yu Haixia, Fu Caikui, Jiang Li. A Fuzzy adaptive tracking algorithm based on currentStatistical Probabilistic Data Association[C].//20102nd international conference onSingal Processing Systems ICSPS,2010: V2-757-759P.
[110] Lihui Wang，Qidan Zhu，Zhouyi Xing. Adaptive Nonlinear Filter Algorithm based onCurrent Statistical Model [C].//Proceedings of the2007International Conference onMechatronics and Automation2007:2414-2418P.
[111] Junliang Chen, Xiaodong Hou, Zheng Qin. A Novel Adaptive Estimator forManeuvering Target tracking[C].//Proceedings of the2007IEEE InternationalConference on Mechatronics and Automation2007:3756-3760P.
[112]李国勇.神经模型控制理论及应用[M].北京:电子工业出版社,2009.
[113]孙增圻.智能控制理论与技术[M].北京:清华大学出版社，1997.
[114]潘平俊,冯新喜,李峰等.基于强跟踪滤波器的自适应常加速模型及跟踪算法[J].系统工程与理论实践,2008,5(5):136-141页.
[115] Pingjun Pan, Xinxi Feng, Fei Li. Self-adaptive Constant Acceleration Model and ItsTracking Alogrithm Based on STF [C].//Proceeding of2007IEEE internationalConference on Mechatronics and Automation, Harbin,2007:3784-3786P.
[116]徐毓,金以慧,杨瑞娟.基于强跟踪滤波器的多目标跟踪方法[J].传感器技术，2002,31(2):17-20页.
[117]冯志全,孟祥旭,蔺永政. UKF滤波器的强跟踪性研究[J].小型微型计算机系统，2006,27(11):2142-2145页.
[118]罗笑冰,王宏强,黎湘等.机动目标跟踪α-Jerk模型[J].信号处理,2007,23(4)：481-484页.
[119]乔向东,王宝树.高度机动目标的“当前”统计Jerk模型[J].系统工程与电子技术,2002,16(24),53-56页.
[120]李大治,李刚,辛勤等.一种基于Jerk模型的机动目标单站无源定位算法[J].国防科技大学学报,2007,4(29):82-86页.
[121]杨翠芳,汪圣利,周丰.基于参数自适应Jerk模型的IMMUKF算法[J].通信与信息处理,2010,29(1):47-52页.
[122] Wenfei Gong，Mingdi Yi，Jianqi Li. An Improved Highly Maneuvering Target Modeland Adaptive Tracking Algorithm [C].//2009IET International Conference on Radar,2010:1-4P.
[123]雷明,韩崇昭.多级修正的高机动Jerk模型研究[J].西安交通大学学报,2006,40(2):138-141页.
[124]宋强,何友,杨俭.基于强跟踪滤波器的Jerk模型目标跟踪算法[J].海军航空工程学院学报,2007,22(3)：329-332页.
[125]乔向东,王宝树,李涛等.一种高度机动目标的当前统计Jerk模型[J].西安电子科技大学学报,2002,29(4):534-539页.
[126]李大治,程翥,皇甫堪.基于Jerk模型的单站无源定位跟踪算法研究[J].系统仿真学报,2008,20(24):6750-6753页.
[127]陈亮,吴小俊,王士同等.基于Jerk的常增益目标跟踪及其自适应算法[J].计算机工程与应用,2008,44(31):72-74页.
[128] Crassidis J L, Markley F L. Predictive Filtering for Nonlinear Systems [J]. Journal ofGuidance, Control and Dynamics,1997,20(3):556-572P.
[129]王小旭,赵琳.基于MEP-UKF的组合导航滤波算法[J].宇航学报,2010,31(2):432-439页.
[130]杨静,张洪钺.预测滤波器理论在惯导非线性对准中的应用[J].中国惯性技术学报,2003,11(6):44-52页.
[131]张红梅,邓正隆,林玉荣.一种基于模型误差预测的UKF方法[J].航空学报,2004,25(6):598-601页.
[132]张惟,林宝军.改进粒子滤波与预测滤波相结合的单星敏姿态估计[J].控制与决策,2011,26(5):655-660页.
[133]曹娟娟,房建成,盛蔚.大失准下MIMU空军快速对准技术[J].航空学报,2007,28(6):1395-1400页.
[134]张惟,林宝军.改进粒子滤波与预测滤波相结合的单星敏姿态估计[J].控制与决策,2011,26(5):655-660页.
[135]杨静,冀红霞,魏明坤.一种非线性滤波器的误差分析及应用[J].航空学报,2011,32(8):1-8页.
[136]吴盘龙,孔建寿.基于平方根UKF的水下纯方位目标跟踪[J].南京理工大学学报(自然科学版),2009,33(6):751-755页.
[137]张树春,胡广大.平方根滤波及其在目标跟踪方面的应用[J].哈尔滨工业大学学报,2008,40(5):700-704页.
[138]郝燕玲,杨峻巍,陈亮等.基于平方根中心差分卡尔曼滤波的大方位失准角初始对准[J].中国惯性技术学报,2011,19(2):211-215页.
[139]谢凯,周一宇,薛模根等.基于平方根UKF的自由段目标跟踪算法[J].国防科学技术大学学报,2007,29(7):97-100页.
[140]陈阳舟,孙章固,马海波.基于平方根UKF的车辆组合导航[J].系统工程与电子技术,2008,30(5):926-928页.
[141]罗笑冰,王宏强,黎湘等.机动目标跟踪MS模型[J].系统工程与电子技术,2006,28(6):813-816页.
[142]吴富梅,杨元喜.基于高阶AR模型的陀螺随机漂移模型[J].测绘学报,2007,36(11):389-394页.
[143]张伟,胡昌华,焦李成.最小二乘AR模型的惯性器件故障预测[J].仪器仪表学报,2006,37(6):1755-1757页.
[144]甘敏,彭辉,陈晓红. RBF-AR模型在非线性时间预测中的应用[J].系统工程理论与实践,2010,30(6):1055-1061页.
[145] X. Rong Li, Zhanlue Zhao, Xiaobai Li. General Model-Set Design Methods forMultiple-Model Approach [J]. IEEE Transaction on Automatic Control,2005,50(9):1260-1276P.
[146] H.A.P.Blom, Y.Bar-Shalom. The interacting multiple model algorithm for systemswith Markovian switching coefficients [J]. IEEE Trans. Automatic Control,1988,33:780-783P.
[147]徐毓,金以慧.相关噪声下的机动目标跟踪.SMFEKF-IMM算法[J].清华大学学报,2003,43(7)865-868页.
[148]肖卫东,聂晓华,郑剑波等.基于转弯模型的自适应交互式多模型跟踪算法[J].指挥控制与仿真,2009,(4):36-41页.
[149]曲彦文,张二华,杨静宇.基于模型误差的交互式多模型算法[J].计算机科学,2008,35(5):134-137页.
[150]曲彦文,张二华,杨静宇.基于UKF与模型误差的交互式多模型算法[J].系统学报,2009,21(12):3565-3569页
[151]梁彦,程咏梅,贾宇岗.交互式多模型算法性能分析[J].控制理论与应用,2001,18(4):487-492页.
[152]孙福明,吴秀清,王鹏伟.转弯机动目标的两层交互式多模型跟踪算法[J].控制理论与应用,2008,25(2):233-236页.
[153]董莎莎,徐一兵,李勇等.双Markov多假设IMM机动目标跟踪算法[J].计算机工程,2010,36(7):204-205页.
[154] Zhanlue Zhao, X. Rong Li. The Behavior of Model Probability in Multiple ModelAlgorithms[C].//20057th International Conference on Information Fusion,2005:331-336P.
[155]徐肖豪,高彦杰,杨国庆.交互式多模型算法中模型集选择的分析研究[J].航空学报,2004,25(4):352-356页.
[156]刘士建,郭立,李士民.变结构多模型估计单Kalman滤波跟踪机动目标算法[J].电路与系统学报,2003,8(5):28-31页.
[157]樊国创,戴亚平,刘岩.模型集自适应的交互多模型辅助粒子滤波算法[J].北京理工大学学报,2008,28(12):1070-1073页.
[158] Jian Lan, X.Rong Li. Equivalent-Model Augmentation for Variable-StructureMultiple-Model Estimation [C].//2011Proceedings of the14th InternationalConference on Information Fusion (FUSION)，2011:1-8P.
[159] Xiangan Li, Zheng Qin, Ronghua Guo.ect. A Variable-Structure IMM FilteringAlgorithm Based on Current Statistical Model[C].//Proceeding of2009IEEEInternational Conference on Mechatronic and Automation, Changchun,2009:2983-2987P.
[160] Jian Lan, X.Rong Li, Chundi Mu. Best Model Augmentation for Variable-StructureMultiple-Model Estimation [J]. IEEE Transactions on Aerospace and ElectronicSystems,2011,47(3):2008-2024P.
[161]高羽,张建秋,尹建君.机动目标的多项式预测模型及其跟踪算法[J].航空学报,2009,30(8):1479-1489页.
[162] Heinonen P, Neuvo Y. FIR-median hybrid filters with predictive FIR substructures [J].IEEE Trans on Acoustics, Speech, and Singal Processing,1988,36(6):892-899P.
[163] Huiming Chen, X. Rong Li, Yaakov Bar-Shalom. On Joint Track Initiation andParameter Estimation under Measurement Origin Uncertainty [J]. IEEE Transactionson Aerospace and Electronic Systems,2004,4(20):675-694P.
[164] Vesselin P. Jilkov, X. Rong Li. Online Bayesian Estimation of Transition Probabilitiesfor Markovian Jump Systems [J]. IEEE Transaction on Singal Processing,2004,52(6):1620-1630P.
[165] Mengjun Jin，Shaohua Hong，Zhiguo Shi,ect. Current statistical model probabilityHypothesis Density Filter for Multiple Maneuvering Targets Tracking [C].//International conference on wirless communications&signal processing,2009:1-5P.
[2] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part II: BallisticTargets Models [C].//Proceeding of SPIE Conference on Singal and Data Processingof Small Targets.San Diego, CA, USA,2001(4473-63):1-23P.
[3] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part III:Meauserment Models [C].//Proceeding of SPIE Conference on Singal and DataProcessing of Small Targets.San Diego, CA, USA,2001(4473-41):1-24P.
[4] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part IV:Decision-Based Methods [C].//Proceeding of SPIE Conference on Singal and DataProcessing of Small Targets.Orlando, FL, USA,2002(4428-60):1-24P.
[5] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part V:Multiple-Model Methods [J]. Submited to IEEE Transaction on Aerospace andElectronic Systems,2003:1-58P.
[6]权太范.目标跟踪新理论与技术[M].北京:国防工业出版社,2009.
[7]潘泉,梁彦,杨峰等.现代目标跟踪与信息融合[M].北京:国防工业出版社,2009.
[8]何友,修建娟,张晶炜等.雷达数据处理及应用(第二版)[M].北京:电子工业出版社,2009.
[9]杨元喜.自适应动态导航定位[M].北京:测绘出版社,2006.
[10]杨峰.现代目标跟踪与多传感器融合关键技术研究[D].西北工业大学工学博士学位论文,2006
[11]夏佩伦编著,目标跟踪与信息融合[M].北京:国防工业出版社,2010.
[12]梁彦,潘泉,杨峰等.复杂系统的现代估计理论及应用[M].北京:科学出版社,2009.
[13]何友,王国宏,陆大紟等.多传感器信息融合及应用(第二版)[M].北京:电子工业出版社,2007.
[14]周宏仁,敬忠良,王培德.机动目标跟踪[M].北京:国防工业出版社,1991.
[15]彭东亮,文成林,薛安克.多传感器多源信息融合理论及应用[M].北京:科学出版社,2010.
[16]石章松,刘忠.目标跟踪与数据融合理论及方法[M].北京：国防工业出版社,2010.
[17]史玉柱.最优估计的计算方法[M].北京:科学出版社,2001.
[18] Jazwinski A H. Stochastic processes and filtering theory [M]. New York: Academic,1970.
[19] Maybeck P S. Stochastic Models，Estimation and Control [M]. New York: AcademicPress,1979.
[20] Sorenson H W. Kalman filtering: Theory and Application [M]. New York: IEEE Press,1985.
[21]俞济祥.卡尔曼滤波及其在惯性导航中的应用[M].北京:航空专业教材编审组,1984.
[22]秦永元,张洪钺,汪叔华.卡尔曼滤波与组合导航原理[M].西安：西北工业大学出版社,1998.
[23]付梦印,邓志红,闫莉萍. Kalman滤波理论及其在导航系统中的应用(第二版)[M].北京：科学出版社,2010.
[24] Dan Simon. Optimal State Estimation [M]. John Wiley&Sons, Inc., Hoboken, NewJersey,2006.
[25] Julier S J, Uhlmann J K. A new approach for filtering nonlinear system[C].//Proceedings of the1995American Control Conference,1995:1628-1632P.
[26] Julier S J, Uhlmann J K. A new method for the nonlinear transformation of means andcovariances in filters and estimators [J]. IEEE Transactions on Automatic Control,2000,45(3):477-482P.
[27] N rgarrd M, Poulsen N K, Ravn O. New developments in state estimation fornonlinear systems [J]. Automatica,2000,36(11):1627-1638P.
[28] Merwe R V D. Sigma-Point Kalman Filters for Probabilistic Inference in DynamicState-Space Models [J]. http://www.cslu.ogi.edu/publications,2004.
[29] Julier S J, Uhlmann J K. A general method for approximating nonlineartransformation of probability distributions [J]. http://www.eng.ox.ac.uk,1996.
[30] Julier S J, Uhlmann J K. Reduced sigma point filters for the propagation of means andcovariances through nonlinear transformations [C].//Proceedings of AmericanControl Conf. Jefferson City,2002:887-892P.
[31] Merwe R V D, Wan E A. The Square-root Unscented Kalman Filter for State andParameter-eatimation [C].//International Conference on Acoustics, Speech, andSignal Processing. Salt Lake City,2001:3461-3464P.
[32] I. Arasaratnam, S. Haykin. Cubature Kalman filters [J]. IEEE Trans AutomaticControl.2009,54:1254-1269P.
[33] Ienkaran Arasaratnam, Simon Haykin. Cubature Kalman Filtering forContinuous-Discrete Systems: Theory and Simulations [J]. IEEE Transactions onSignal Processing,2010,58:4977-4993P.
[34] Ienkaran Arasaratnam, Simon Haykin. Cubature Kalman smoothers [J].Automatica(2011),doi:10.1016/j.automatica.2011.8.005
[35]兰剑,慕春棣.基于参考加速度的机动目标跟踪模型[J].清华大学学报,2008,48(10):1549-1552页.
[36]肖雷,刘高峰,魏建仁.几种机动目标运动模型的跟踪性能对比[J].火力与指挥控制,2007,(5):106-109页.
[37] Su Y P. The Research on the Maneuvering Object base on Adaptive followingarithmetic[C].//2010International Conference on Intelligent Computation Technologyand Automation,2010:938-941P.
[38] R.L.T.Hampton, J.R.Cooke. Unsupervised tracking of maneuvering vehicles [J], IEEETransaction on Aerospace and Electronic Systems,1973,9(2):197-297P.
[39] Friedland B. Optimum steady-state position and velovity estimation using noisysampled position data [J]. IEEE Transaction on Aerospace and Electronic Systems,1973,9(6):906-111P.
[40] Signger R A. Estimating optimal tracking filter performance for manned maneuveringtargets [J]. IEEE Trans. on Aerospace and Electromic Systems,1970,6(7):473-183P.
[41] R.L.Moose. An adaptive state estimation solution to the maneuvering target problem[J]. IEEE Transactions on Automatic Control,1975,20(6):359-362P.
[42] Zhuo H, Kumar K S P. A “current” statistical model and adaptive algorithm forestimating maneuvering targets [J]. AIAA Journal of Guidance,1984,7(5):596-602P.
[43]张伯彦,蔡庆宇,袁曾任.一种跟踪机动目标的新方法[J].系统工程与电子技术，1990,12(1):45-55页.
[44] Mehrotra K, Mahapatra P R. A Jerk Model for Tracking Highly Maneuvering Targets[J]. IEEE Trans. on Aerospace and Electronic Systems,1997,33(4):1094-1104P
[45] S. Koteswara Rao. Comments on “A Jerk model for highly maneuvering Targets”[J].IEEE transaction on aerospace and electronic system,1998,34(3):982-983P.
[46] Pravas R.Mahapatra. Kishore Mehrotra Mix coordinates tracking of generalizedmaneuvering targets using acceleration and Jerk models [J]. IEEE transaction onaerospace and electronic system,2000,36(3):992-1000P.
[47] GA.Waston, W.D.Blair. IMM algorithm for tracking targets that maneuvering throughcoordinated turns[C].//Proceedings of SPIE Signal and Data Processing of SmallTargets,1992:236-247P.
[48] R.A.Best, J.P.Norton. A new model and efficient tracker for a target with curvilinearmotion [J]. IEEE transaction on aerospace and electronic system,1997,33(31):1030-1037P.
[49] T.Matsuzaki, H.Kameda, S.Tsujimichi.ect. Maneuvering target tracking using constantvelocity and constant angular velocity model[C].//SICE’99,1999:1135-1138P.
[50] J.P.Herferty. Maneuvering Targets: The Use of Turn-rate Distributions for AccretionModeling [J].IEEE Trans. on Aerospace and Electronic Systems,1996,32(4):1355-1361P.
[51] X. Rong Li, Vesselin P. A Survey of Maneuvering Target Tracking. Part II: BallisticTargets Models[C].//Proceeding of SPIE Conference on Singal and Data Processingof Small Targets.San Diego, CA, USA,2001(4473-63):1-23P.
[52]范小军,刘峰.一种新的机动目标跟踪的多模型算法[J].电子与信息学报,2007,29(3):532-535页.
[53]曹梦龙,崔平远.多模型高精度组合导航算法研究[J].系统工程与电子技术，2008,30(7):1304-1307页.
[54]梁彦,程咏梅,贾宇岗.交互式多模型算法性能分析[J].控制理论与应用,2001,18(4):487-492页.
[55]黄翔宇,彭冬亮.一种变结构多模型机动目标跟踪算法的研究[J].杭州电子科技大学学报,2010,30(4):195-199页.
[56]李俊,周凤歧,周军.机动目标跟踪的多模型多机动策略算法[J].系统仿真学报，2009,21,(3):668-671页.
[57]马海平,陈子栋.一种基于H∞滤波的模糊变结构交互式多模型算法[J].电子学报，2008,36(2):245-249页.
[58]徐肖豪,高彦杰,杨国庆.交互式多模型算法中模型集选择的分析研究[J].航空学报,2004,25(4):352-356页.
[59] X. Rong Li, Vesselin P. Jilkov. Multiple-Model Estimation with Variable Structure [J].IEEE Transaction on Automatic Control,1996,41(4):478-493P.
[60] X. Rong Li. Multiple-Model Estimation with Variable Structure II: Model-SetAdaptation [J]. IEEE Transaction on Automatic Control,2000,45(11):2047-2060P.
[61] X. Rong Li, Xiaorong Zhi, Youmin Zhang. Multiple-Model Estimation with VariableStructure III: Model-Group Switching Algorithm [J]. IEEE Transactions on Aerospaceand Electronic Systems,1999,35(1):225-241P.
[62] X. Rong Li, Youmin Zhang, Xiaorong Zhi. Multiple-Model Estimation with VariableStructure IV: Design and Evaluation of Model-Group Switching Algorithm [J]. IEEETransactions on Aerospace and Electronic Systems,1999,35(1):242-254P.
[63] X. Rong Li, Youmin Zhang. Multiple-Model Estimation with Variable Structure V:Likely-Model Set Algorithm [J]. IEEE Transactions on Aerospace and ElectronicSystems,2000,36(2):448-466P.
[64] X. Rong Li, Vesselin P. Jilkov, Jifeng Ru. Multiple-Model Estimation with VariableStructure VI: Expected-Mode Augmentation [J]. IEEE Transactions on Aerospace andElectronic Systems,2005,41(3):853-867P.
[65] X.Rong Li, Vesselin P.Jilkov. Survey of Maneuvering Target Tracking Part II: MotionModels of Ballistic and Space Targets [J]. IEEE Transactions on Aerospace andElectronic Systems,2010,46(1):96-119P.
[66] Jaipal R. Katkuri, Vesselin P. Jilkov, X. Rong Li. A Comparative Study of NonlinearFilters for Target Tracking in Mixed Coordinates [C].//42nd South EasternSymposium on System Theory Tyler, TX, USA,2010:202-207P.
[67] Huiming Chen, X. Rong Li, Yaakov Bar-Shalom. On Joint Track Initiation andParameter Estimation under Measurement Origin Uncertainty [J]. IEEE Transactionson Aerospace and Electronic Systems,2004,4(20):675-694P.
[68]杨柏胜.被动多传感器探测目标跟踪技术研究[D].西安电子科技大学工学博士学位论文,2009.
[69]彭亮.机动目标跟踪算法研究[D].西北工业大学工学博士学位论文,2007.
[70]张高煜.目标跟踪与多传感器信息融合若干问题研究[D].西安电子科技大学工学博士学位论文,2006.
[71]乔向东.信息融合系统中目标跟踪技术[D].西安电子科技大学工学博士学位论文,2003.
[72]孙福明.机动目标跟踪状态估计与数据关联技术研究[D].中国科学技术大学工学博士学位论文,2007.
[73]罗笑冰.强机动目标跟踪技术研究[D].国防科学技术大学工学博士学位论文,2007.
[74] Smith A F M, Geland A E. Bayesian Statistics without tears: a sampling-resamplingperspective [J]. The American Statistican.1992,46(2):84-88P.
[75] Haykin S, Huber K, Chen Z. Bayesian sequential state estimation for MIMO wirelesscommunications [C].//Proceedings of the IEEE,2004,92(3):439-454P.
[76] Handschin J E. Monte Carlo techniques for prediction and filtering of nonlinearstochastic processes [J]. Automatica,1970,6(3):555-563P.
[77]王小旭.非线性SPKF滤波算法研究及其在组合导航中的应用[D].哈尔滨工程大学工学博士学位论文,2010.
[78]郝钢.自校正观测融合Kalman估值器及其在典型跟踪系统中的应用[D].哈尔滨工程大学工学博士学位论文,2011.
[79]李涛.非线性滤波方法在导航系统中的应用研究[D].国防科技大学工学博士学位论文,2003.
[80]杨萌.非线性滤波及其在惯性导航系统传递对准中应用研究[D].哈尔滨工程大学工学博士学位论文,2010.
[81]聂琦.非线性滤波及其在导航系统中的应用[D].哈尔滨工程大学工学博士学位论文,2008.
[82]尹建军.线性/非线性系统的混合动态滤波理论及应用[D].复旦大学工学博士学位论文,2008.
[83] Weisheng Wu, Chunlei Song, Junhou Wang. Cubature Gaussian Particle Filter ForInitial Alignment of Strapdown Inertial Navigation System [C].//2010FirstInternational Conference on Pervasive Computing, Signal Processing andApplications,2010:1196-1200P.
[84] H. Pesonen, R. Pich′e. Cubature-based Kalman Filters for Positioning [C].//20107thWorkshop on Positioning Navigation and Communication (WPNC),2010:45-49P.
[85] Davide Macagnano, Giuseppe Thadeu Freitas de Abreu. Multitarget Tracking with theCubature Kalman Probability Hypothesis Density Filter [J].2010Conference Recordof the Forty Fourth Asilomar Conference on Signals, Systems and Computers(ASILOMAR),2010:1455-1459P.
[86] Kumar Pakki. Bharani Chandra, Da-Wei Gu.ect. Cubature Information Filter and itsApplications[C].//2011American Control Conferenceon O'Farrell Street, SanFrancisco, CA, USA,2011:3609-3614P.
[87] Jing Mu, Yuan-li Cai. Iterated Cubature Kalman Filter and its Application[C].//Proceedings of the2011IEEE International Conference on Cyber Technology inAutomation, Control, and Intelligent Systems, Kunming, China,2011:33-37P.
[88] Chen Hai, Shan Ganlin. Attitude Angle Aided IMMCKF Algorithm [C].//The TenthInternational Conference on Electronic Measurement&Instruments,2011:197-200P.
[89]郝利超,雷虎民,王飞等.基于当前统计模型的模糊自适应跟踪算法[J].电光与控制,2009,16(10):18-21页.
[90]任少伟,王睿,张平定.基于机动频率自适应的目标跟踪算法[J].空军工程大学学报(自然科学版),2004,5(5):32-35页.
[91]王军政,沈伟,赵红波.机动目标跟踪中机动频率的自适应调整[J].北京理工大学学报,2007,27(1):38-41页.
[92]陈佳俊,刘高峰,辛晋生等.机动频率模糊自适应目标跟踪算法研究[J].弹箭与制导学报,2010,30(2):259-262页.
[93]刘高峰,顾雪峰,王华楠.跟踪反舰导弹的机动频率自适应算法[J].计算机仿真，2008,25(10):59-61页.
[94]刘昌云,刘进忙,陈长兴等.机动目标跟踪的机动频率自适应算法[J].控制理论与应用,2004,21(6):961-965页.
[95] WANG W, HOU H L. An improved Current Statistical Model for maneuvering targettracking[C].//4th IEEE Conference on Industrial Electronics and Applications,2009:4017-4020P.
[96] Wang Wei， Huo Hong-lu. An improved Current Statistical Model for maneuveringTargets Tracking[C].//4th IEEE Conference on Industrial Electronics andApplications,2009:4017-4020P.
[97]王炜,杨露菁.一种改进的当前统计模型及其自适应滤波算法[J].系统仿真学报,2005,17(8):2037-2040页.
[98]李辉,沈莹,张安等.机动目标跟踪中一种新的自适应滤波算法[J].西北工业大学学报,2006,24(3):354-357页.
[99]段战胜,韩崇昭,党宏社.最大加速度未知的当前统计模型机动目标跟踪[J].计算机工程与应用,2003,(29):19-22页.
[100]巴宏欣,赵宗贵,杨飞等.机动目标的模糊自适应跟踪算法[J].系统仿真学报,2004,16(6):1181-1186页.
[101]王炜,杨露菁.一种改进的当前统计模型及其自适应滤波算法[J].系统仿真学报,2005,(3):2037-2040页.
[102]胡洪涛,敬忠良,田宏伟等.基于“当前”统计模型的模糊自适应跟踪算法[J].系统仿真学报,2005,17(2):293-299页.
[103]王向华,覃征,杨慧杰等.基于“当前”统计模型的模糊自适应跟踪算法[J].兵工学报,2009,(8):1089-1093页
[104]刁联旺,杨静宇.一种改进的机动目标“当前”统计模型的描述[J].兵工学报,2005,26(6):625-628页.
[105]胡振涛,刘先省.基于当前统计模型的一种改进机动目标跟踪算法[J].山东大学学报(工学版),2005,35(3):111-114页.
[106] Xiaoxiao Wu，Fuchun Sun，Wansen Wang.ect. Adaptive algorithm of trackingmaneuvering targets using rate of acceleration [C].//Proceeding of the6th WorldCongress on Intelligent Control and Automation Dalian,2006:1838-1841P.
[107] Yu H X, Fu C K, Jiang L. A fuzzy adaptive tracking algorithm based on currentstatistical probabilistic data association[C].//2nd International Conference on SignalProcessing Systems,2010:757-759P.
[108] Wang L H, Zhu Q D, Xing Z Y. Adaptive Nonlinear Filter Algorithm Based OnCurrent Statistical Model[C].//Proc. of the2007IEEE International conference onMechatronics and automation, Harbin,2007:2414-2418P.
[109] Yu Haixia, Fu Caikui, Jiang Li. A Fuzzy adaptive tracking algorithm based on currentStatistical Probabilistic Data Association[C].//20102nd international conference onSingal Processing Systems ICSPS,2010: V2-757-759P.
[110] Lihui Wang，Qidan Zhu，Zhouyi Xing. Adaptive Nonlinear Filter Algorithm based onCurrent Statistical Model [C].//Proceedings of the2007International Conference onMechatronics and Automation2007:2414-2418P.
[111] Junliang Chen, Xiaodong Hou, Zheng Qin. A Novel Adaptive Estimator forManeuvering Target tracking[C].//Proceedings of the2007IEEE InternationalConference on Mechatronics and Automation2007:3756-3760P.
[112]李国勇.神经模型控制理论及应用[M].北京:电子工业出版社,2009.
[113]孙增圻.智能控制理论与技术[M].北京:清华大学出版社，1997.
[114]潘平俊,冯新喜,李峰等.基于强跟踪滤波器的自适应常加速模型及跟踪算法[J].系统工程与理论实践,2008,5(5):136-141页.
[115] Pingjun Pan, Xinxi Feng, Fei Li. Self-adaptive Constant Acceleration Model and ItsTracking Alogrithm Based on STF [C].//Proceeding of2007IEEE internationalConference on Mechatronics and Automation, Harbin,2007:3784-3786P.
[116]徐毓,金以慧,杨瑞娟.基于强跟踪滤波器的多目标跟踪方法[J].传感器技术，2002,31(2):17-20页.
[117]冯志全,孟祥旭,蔺永政. UKF滤波器的强跟踪性研究[J].小型微型计算机系统，2006,27(11):2142-2145页.
[118]罗笑冰,王宏强,黎湘等.机动目标跟踪α-Jerk模型[J].信号处理,2007,23(4)：481-484页.
[119]乔向东,王宝树.高度机动目标的“当前”统计Jerk模型[J].系统工程与电子技术,2002,16(24),53-56页.
[120]李大治,李刚,辛勤等.一种基于Jerk模型的机动目标单站无源定位算法[J].国防科技大学学报,2007,4(29):82-86页.
[121]杨翠芳,汪圣利,周丰.基于参数自适应Jerk模型的IMMUKF算法[J].通信与信息处理,2010,29(1):47-52页.
[122] Wenfei Gong，Mingdi Yi，Jianqi Li. An Improved Highly Maneuvering Target Modeland Adaptive Tracking Algorithm [C].//2009IET International Conference on Radar,2010:1-4P.
[123]雷明,韩崇昭.多级修正的高机动Jerk模型研究[J].西安交通大学学报,2006,40(2):138-141页.
[124]宋强,何友,杨俭.基于强跟踪滤波器的Jerk模型目标跟踪算法[J].海军航空工程学院学报,2007,22(3)：329-332页.
[125]乔向东,王宝树,李涛等.一种高度机动目标的当前统计Jerk模型[J].西安电子科技大学学报,2002,29(4):534-539页.
[126]李大治,程翥,皇甫堪.基于Jerk模型的单站无源定位跟踪算法研究[J].系统仿真学报,2008,20(24):6750-6753页.
[127]陈亮,吴小俊,王士同等.基于Jerk的常增益目标跟踪及其自适应算法[J].计算机工程与应用,2008,44(31):72-74页.
[128] Crassidis J L, Markley F L. Predictive Filtering for Nonlinear Systems [J]. Journal ofGuidance, Control and Dynamics,1997,20(3):556-572P.
[129]王小旭,赵琳.基于MEP-UKF的组合导航滤波算法[J].宇航学报,2010,31(2):432-439页.
[130]杨静,张洪钺.预测滤波器理论在惯导非线性对准中的应用[J].中国惯性技术学报,2003,11(6):44-52页.
[131]张红梅,邓正隆,林玉荣.一种基于模型误差预测的UKF方法[J].航空学报,2004,25(6):598-601页.
[132]张惟,林宝军.改进粒子滤波与预测滤波相结合的单星敏姿态估计[J].控制与决策,2011,26(5):655-660页.
[133]曹娟娟,房建成,盛蔚.大失准下MIMU空军快速对准技术[J].航空学报,2007,28(6):1395-1400页.
[134]张惟,林宝军.改进粒子滤波与预测滤波相结合的单星敏姿态估计[J].控制与决策,2011,26(5):655-660页.
[135]杨静,冀红霞,魏明坤.一种非线性滤波器的误差分析及应用[J].航空学报,2011,32(8):1-8页.
[136]吴盘龙,孔建寿.基于平方根UKF的水下纯方位目标跟踪[J].南京理工大学学报(自然科学版),2009,33(6):751-755页.
[137]张树春,胡广大.平方根滤波及其在目标跟踪方面的应用[J].哈尔滨工业大学学报,2008,40(5):700-704页.
[138]郝燕玲,杨峻巍,陈亮等.基于平方根中心差分卡尔曼滤波的大方位失准角初始对准[J].中国惯性技术学报,2011,19(2):211-215页.
[139]谢凯,周一宇,薛模根等.基于平方根UKF的自由段目标跟踪算法[J].国防科学技术大学学报,2007,29(7):97-100页.
[140]陈阳舟,孙章固,马海波.基于平方根UKF的车辆组合导航[J].系统工程与电子技术,2008,30(5):926-928页.
[141]罗笑冰,王宏强,黎湘等.机动目标跟踪MS模型[J].系统工程与电子技术,2006,28(6):813-816页.
[142]吴富梅,杨元喜.基于高阶AR模型的陀螺随机漂移模型[J].测绘学报,2007,36(11):389-394页.
[143]张伟,胡昌华,焦李成.最小二乘AR模型的惯性器件故障预测[J].仪器仪表学报,2006,37(6):1755-1757页.
[144]甘敏,彭辉,陈晓红. RBF-AR模型在非线性时间预测中的应用[J].系统工程理论与实践,2010,30(6):1055-1061页.
[145] X. Rong Li, Zhanlue Zhao, Xiaobai Li. General Model-Set Design Methods forMultiple-Model Approach [J]. IEEE Transaction on Automatic Control,2005,50(9):1260-1276P.
[146] H.A.P.Blom, Y.Bar-Shalom. The interacting multiple model algorithm for systemswith Markovian switching coefficients [J]. IEEE Trans. Automatic Control,1988,33:780-783P.
[147]徐毓,金以慧.相关噪声下的机动目标跟踪.SMFEKF-IMM算法[J].清华大学学报,2003,43(7)865-868页.
[148]肖卫东,聂晓华,郑剑波等.基于转弯模型的自适应交互式多模型跟踪算法[J].指挥控制与仿真,2009,(4):36-41页.
[149]曲彦文,张二华,杨静宇.基于模型误差的交互式多模型算法[J].计算机科学,2008,35(5):134-137页.
[150]曲彦文,张二华,杨静宇.基于UKF与模型误差的交互式多模型算法[J].系统学报,2009,21(12):3565-3569页
[151]梁彦,程咏梅,贾宇岗.交互式多模型算法性能分析[J].控制理论与应用,2001,18(4):487-492页.
[152]孙福明,吴秀清,王鹏伟.转弯机动目标的两层交互式多模型跟踪算法[J].控制理论与应用,2008,25(2):233-236页.
[153]董莎莎,徐一兵,李勇等.双Markov多假设IMM机动目标跟踪算法[J].计算机工程,2010,36(7):204-205页.
[154] Zhanlue Zhao, X. Rong Li. The Behavior of Model Probability in Multiple ModelAlgorithms[C].//20057th International Conference on Information Fusion,2005:331-336P.
[155]徐肖豪,高彦杰,杨国庆.交互式多模型算法中模型集选择的分析研究[J].航空学报,2004,25(4):352-356页.
[156]刘士建,郭立,李士民.变结构多模型估计单Kalman滤波跟踪机动目标算法[J].电路与系统学报,2003,8(5):28-31页.
[157]樊国创,戴亚平,刘岩.模型集自适应的交互多模型辅助粒子滤波算法[J].北京理工大学学报,2008,28(12):1070-1073页.
[158] Jian Lan, X.Rong Li. Equivalent-Model Augmentation for Variable-StructureMultiple-Model Estimation [C].//2011Proceedings of the14th InternationalConference on Information Fusion (FUSION)，2011:1-8P.
[159] Xiangan Li, Zheng Qin, Ronghua Guo.ect. A Variable-Structure IMM FilteringAlgorithm Based on Current Statistical Model[C].//Proceeding of2009IEEEInternational Conference on Mechatronic and Automation, Changchun,2009:2983-2987P.
[160] Jian Lan, X.Rong Li, Chundi Mu. Best Model Augmentation for Variable-StructureMultiple-Model Estimation [J]. IEEE Transactions on Aerospace and ElectronicSystems,2011,47(3):2008-2024P.
[161]高羽,张建秋,尹建君.机动目标的多项式预测模型及其跟踪算法[J].航空学报,2009,30(8):1479-1489页.
[162] Heinonen P, Neuvo Y. FIR-median hybrid filters with predictive FIR substructures [J].IEEE Trans on Acoustics, Speech, and Singal Processing,1988,36(6):892-899P.
[163] Huiming Chen, X. Rong Li, Yaakov Bar-Shalom. On Joint Track Initiation andParameter Estimation under Measurement Origin Uncertainty [J]. IEEE Transactionson Aerospace and Electronic Systems,2004,4(20):675-694P.
[164] Vesselin P. Jilkov, X. Rong Li. Online Bayesian Estimation of Transition Probabilitiesfor Markovian Jump Systems [J]. IEEE Transaction on Singal Processing,2004,52(6):1620-1630P.
[165] Mengjun Jin，Shaohua Hong，Zhiguo Shi,ect. Current statistical model probabilityHypothesis Density Filter for Multiple Maneuvering Targets Tracking [C].//International conference on wirless communications&signal processing,2009:1-5P.