A strong tracking nonlinear robust filter for eye tracking

Non-intrusive methods for eye tracking are important for many applications of vision-based human computer interaction. However, due to the high nonlinearity of eye motion, how to ensure the robustness of external interference and accuracy of eye tracking pose the primary obstacle to the integration of eye movements into today’s interfaces. In this paper, we present a strong tracking unscented Kalman filter (ST-UKF) algorithm, aiming to overcome the difficulty in nonlinear eye tracking. In the proposed ST-UKF, the Suboptimal fading factor of strong tracking filtering is introduced to improve robustness and accuracy of eye tracking. Compared with the related Kalman filter for eye tracking, the proposed ST-UKF has potential advantages in robustness and tracking accuracy. The last experimental results show the validity of our method for eye tracking under realistic conditions.     

用于弹道目标跟踪的新的非线性滤波算法

针对再入阶段弹道目标的跟踪问题,提出一种新的自适应滤波算法,即强跟踪有限差分扩展卡尔曼滤波（STFDEKF）算法,用于非线性系统的目标跟踪。该方法使用Sterling内插公式进行多项式的近似,从而实现对非线性函数的近似,避免了非线性函数的求导运算;并且算法中引入强跟踪的因子来修正先验的协方差矩阵。新算法改进了跟踪精度,扩大了应用范围,增强了滤波收敛性。仿真实验将新算法与扩展卡尔曼滤波器（EKF）、有限差分扩展卡尔曼滤波器（FDEKF）进行了比较,结果表明,STFDEKF在跟踪精度和滤波可靠性上均优于EKF和FDEKF,但其计算复杂性更大。得出结论,STFDEKF是个很有效的非线性滤波算法。     

A new real-time eye tracking based on nonlinear Unscented Kalman Filter for monitoring driver fatigue

A new scheme for driver fatigue detection is presented, which is based on the nonlinear unscented Kalman filter and eye tracking. Assuming a probability distribution than to approximate an arbitrary nonlinear function or transformation, eye nonlinear tracking can be achieved using an unscented transformation (UT), which adopts a set of deterministic sigma points to match the posterior probability density function of the eye movement. Driver fatigue can be detected using the percentage of eye closure (PERCLOS) framework in a realistic driving condition after the eye nonlinear tracking. This system was tested adequately in realistic driving environments with subjects of different genders, with/without glasses, in day/night driving, being commercial/noncommercial drivers, in continuous driving time, and under different road conditions. The last experimental results show that the proposed method not only improves the robustness for nonlinear eye tracking, but also can provide more accurate estimation than the traditional Kalman filter.     

强跟踪平方根容积卡尔曼滤波和 自回归模型融合的故障预测

为了解决非线性系统中不可测量参数的预测问题,提出一种带有次优渐消因子的强跟踪平方根容积卡尔曼滤波(STSCKF)和自回归(AR)模型相结合的故障预测方法.利用AR模型时间序列预测法预测未来时刻的测量值,将预测的测量值作为STSCKF的测量变量,从而将预测问题转化为滤波估计问题.STSCKF通过在预测误差方差阵的均方根中引入渐消因子调节滤波过程中的增益矩阵,克服了故障参数变化函数未知情况下普通SCKF跟踪故障参数缓慢甚至失效的局限性,使得STSCKF能较好地预测故障参数的发展趋势.连续搅拌反应釜(CSTR)仿真结果表明,STSCKF的预测精度高于普通SCKF和强跟踪无迹卡尔曼滤波(STUKF),验证了方法的有效性.     

基于强跟踪滤波器的多目标跟踪方法

在诸多的多目标跟踪算法中，相互作用多模型（IMM）算法是目前公认的最为有效的算法。但到目前为止，LMM估计方法都是建立在卡尔曼滤波器（KF）和扩展卡尔曼滤波器（EKF）基础上，因而其性能不仅依赖于所采用的模型集，而且在更大程度上依赖于所采用的滤波技术。强跟踪滤波器（STF）克服了卡尔曼和扩展卡尔曼的三大缺陷，因而设计一种基于STF的IMM目标跟踪算法显然能提高其性能。仿真实验表明，基于STF的IMM算法的跟踪性能要优于基于KF和EKF的IMM算法的跟踪性能。     

基于加性无迹卡尔曼滤波的雷达目标跟踪方法

提出一种基于加性无迹卡尔曼滤波的雷达目标跟踪方法。雷达跟踪系统为离散非线性系统,传统的解决方法是使用扩展卡尔曼滤波。无迹卡尔曼滤波用少量采样点表示随机变量的分布,通过非线性系统传播,能以三阶精度获得非线性变换的均值和方差的估计。用无迹卡尔曼滤波进行雷达目标跟踪。通过Monte Carlo仿真,验证了该滤波算法比传统的扩展卡尔曼滤波具有更高的滤波精度。     

Observer-based optimal output tracking for discrete-time systems with multiple state and input delays

The optimal output tracking control (OOTC) problem for a class of discrete-time systems with state and input delays is addressed. An augmented system is constructed such that the OOTC problem can be transformed into a two-point boundary value (TPBV) problem with both advance and delay terms from the necessary optimality conditions. The successive approximating method recently developed is extended to obtain an approximate solution of the TPBV problem, which is then used to obtain a feedforward and feedback tracking controller. An observer is designed for the uncertain reference input such that the feedforward controller is physically realizable. Simulations show the results are effective even with long time-delays. Recommended by Editorial Board member Poo Gyeon Park under the direction of Editor Young Il Lee. This research was supported by the National Natural Science Foundation of China (Grant No. 40776051), the Key Natural Science Foundation of Shandong Province (Grant No. Z2005G01), the Natural Science Foundation of Qingdao City (Grant No. 05-1-JC-94) and the research funds of QingDao University of Science and Technology. Hai-Hong Wang received the Ph.D. degree in Computer Science in July 2007 from Ocean University of China. She presently works in QingDao University of Science and Technology, Qingdao, P.R. China. Her current research interests include analysis and control for time-delay systems and nonlinear systems. Gong-You Tang received the Ph.D. degree in Control Theory and Applications from the South China University of Technology, P. R. China in 1991. He is a Professor at the College of Information Science and Engineering at the Ocean University of China, Qingdao, P. R. China. He is the Editor of the Journal of the Ocean University of China and Control and the Instruments in Chemical Industry. His research interests are in the areas of nonlinear systems, delay systems, large-scale systems, and networked control systems, with emphasis in optimal control, robust control, fault diagnosis and stability analysis.     

适用处理非高斯观测噪声的强跟踪卡尔曼滤波器

在高斯噪声条件下,卡尔曼滤波器(KF)能够获得系统状态的一致最小方差线性无偏估计.但当噪声非高斯,KF性能将严重下降.观测噪声非高斯现象在深空探测自主导航中经常遇到,然而现有模型可能存在着精度不高、稳定性不强或者计算复杂度较高的缺点.针对这种现状,本文在传统强跟踪卡尔曼滤波器(STKF)中新息正交原则的基础上,推导了适用处理非高斯观测噪声的强跟踪卡尔曼滤波器(STKFNO),并将其嵌入到无迹卡尔曼滤波(UKF)框架下形成适用处理非线性系统非高斯观测噪声的强跟踪无迹卡尔曼滤波器(STUKFNO).所提出的算法被应用到深空光学自主导航系统中,仿真结果表明所提出的算法能够较好地应对观测噪声的非高斯性.     

Target tracking based on the extended H-infinity filter in wireless sensor networks

In this paper, we propose a new target tracking approach for wireless sensor networks (WSNs) by using the extended H-infinity filter. First, the extended H-infinity filter for nonlinear discrete-time systems is deduced through the Krein space analysis scheme. Then, the proposed extended H-infinity filtering algorithm is applied to target tracking in wireless sensor networks. Finally, experiments are conducted through a small wireless sensor network test-bed. Both experimental and simulation results illustrate that the extended H-infinity filtering algorithm is more accurate to track a moving target in wireless sensor networks than using the extended Kalman filter in the case of having no knowledge of the statistics of the environment and the target to be tracked.     

LOS/NLOS混合环境中基于交互式多模型的鲁棒目标跟踪

针对视距（line-of-sight, LOS）和非视距(none-line-of-sight, NLOS)混合环境的定位跟踪问题, 提出一种基于扩展H∞滤波（EHF）和扩展卡尔曼滤波(EKF)的交互式多模型(IMM)定位跟踪算法。目标与基站之间的LOS、NLOS传输信道分别用EKF和EHF匹配, 模型间的转换用马尔可夫过程进行描述。Monte Carlo仿真结果表明, 该算法具有较高的定位精度、较好的跟踪稳定性, 且计算时间与基于EKF的IMM算法相当, 具有较好的可实现性。     

强跟踪卡尔曼滤波在视频目标跟踪中的应用

针对经典卡尔曼滤波器在滤波数学模型与实际过程的数学规律不匹配、滤波特性较差的情况,提出利用强跟踪卡尔曼滤波器对视频序列图像中的运动目标进行跟踪。该方法是在经典卡尔曼滤波递推公式中的一步验前误差方差阵中引入可在线计算的时变渐消矩阵,从而调节增益K,使之能够不断变化,保证对新息序列的自适应调节,使状态滤波更准确。实验结果表明,较之经典卡尔曼滤波,该方法具有对运动目标更强的跟踪能力,跟踪精度更高,均方误差更小。     

一种强跟踪扩展卡尔曼滤波器的改进算法

针对模型不匹配卡尔曼的状态估计发散和应用范围限于连续系统问题,提出一种基于有限差分强跟踪滤波嚣（STFDEKF）．在滤波计算中,引入强跟踪滤波因子修正滤波器的状态预协方差矩阵。滤波精度得以提高,滤波器应用有限差分方法计算滤波过程中非线性函数的偏导数。扩大了适用范围．几种卡尔曼滤波器经过仿真比较。STFDEKF应用于复杂非线性系统状态估计时．具有较高数值稳定性、强跟踪性和较宽应用范围．     

基于双眼同步运动特征约束的Kalman瞳孔跟踪算法

针对注视点研究中,红外光照明下双眼瞳孔运动的定位跟踪存在误差的问题,以双眼实时图像为研究对象,提出一种基于双眼同步运动特征约束的瞳孔跟踪算法.根据人类双眼在注视过程中的同步运动特征,把双眼瞳孔间距矢量作为隐式参数进行估计,简化包含左右眼位置、速度和双眼瞳孔间距的模型为统一的双眼同步跟踪模型,运用Kalman滤波器实现了运动特征估计和状态跟踪.实验采用自制的头戴式注视点传感装置进行眼部图像的采集.实验表明,该算法跟踪精度高,抗干扰能力强.相较于传统的以左右瞳孔位置与速度以及左右眼相对位置为状态量的算法,本文算法在位置跟踪和速度跟踪的鲁棒性上有明显改善,算法计算量也明显减少.另外,经过本文算法处理后的注视点位置估计精度大大提高,为注视点在人机交互领域中的进一步应用奠定了基础.     

改进扩展卡尔曼滤波算法的目标跟踪算法

为了获得更加理想的运动目标跟踪效果,提出了一种基于改进扩展卡尔曼滤波的目标跟踪算法。构建时间差和信号到达方向的观测方程,利用几何和代数关系化简得到伪线性模型,通过改进卡尔曼滤波算法对目标运动轨迹进行跟踪,采用仿真实验对算法性能进行测试。结果表明,相对于传统扩展卡尔曼滤波算法,在相同条件下,该算法不仅提高了目标跟踪精度,而且使目标跟踪结果更加稳定。     

分布式动态一致性非线性目标跟踪策略研究

针对集中目标跟踪和分层目标跟踪中心节点通信瓶颈以及容错性能差的不足, 提出了一种分布式动态一致性非线性目标跟踪策略。目标状态初始化由网络节点采用加权最小二乘法完成。整个跟踪过程采用动态成簇策略, 分阶段选择并唤醒任务节点检测目标并执行分布式一致性扩展卡尔曼滤波策略完成目标的状态估计, 其余节点进入休眠状态从而能降低系统的能耗。从跟踪误差和能量两个方面, 与集中目标跟踪算法相比, 仿真结果表明所提算法与集中卡尔曼滤波相比, 跟踪精度相当, 适用于要求高可靠度的非线性跟踪。此外分布式的工作方式使得节点仅需与邻居交换数据并在局部完成状态估计, 消除集中式结构中心节点的瓶颈, 以保证部分传感器节点的损坏不会影响到全局任务的完成。     

一种新的非线性目标跟踪方法

在实际的目标跟踪场景中,普遍存在非高斯过程噪声和/或量测噪声,以及非高斯先验信息等情况,针对这一问题,提出一种新的解决非线性/非高斯系统滤波问题的非线性滤波算法,即高斯和求积分卡尔曼滤波（GSQKF）算法。仿真实验将新算法与标准的粒子滤波算法进行了比较,表明新算法是一种非常有效的非线性滤波算法。     

基于强跟踪滤波器的MACA-MIE 模型及跟踪算法

结合自适应常加速模型(ACA)、改进输入估计(MIE)和强跟踪滤波器,提出一种新的自适应目标跟踪模型和算法.该算法通过扩展 ACA 模型状态矢量和改进状态噪声协方差调整方法,利用 MIE 和强跟踪滤波器,实现了机动加速度方差和状态预测协方差依据残差信息的实时完全自适应调整,在缺乏目标加速度先验知识的情况下,能够实时高精度跟踪目标突变状态、弱机动和非机动状态.仿真实验表明,相比 ACA 模型和 MIE,该算法具有更好的机动状态和非机动状态跟踪性能.     

改进的KCF算法在车辆跟踪中的应用

针对核相关滤波算法(KCF)在复杂道路场景下难以应对因车辆尺度变化,遮挡及旋转而不能继续跟踪的问题,提出了一种新的跟踪方法来更好地实现复杂道路场景下的车辆跟踪。该方法借鉴快速分类尺度空间跟踪器(fDDST),采用一维尺度相关滤波器进行尺度估计。同时融合Kalman滤波器形成预测-跟踪-校准的跟踪机制。该机制结合遮挡处理能够保证系统在目标被严重遮挡时跟踪的准确性。在模型更新方面,在目标被遮挡时,自适应的调节学习率参数,及时纠正模型偏移、特征丢失等问题。实验结果表明,在复杂道路场景下车辆旋转 、遮挡及尺度变化时,均能有效地跟踪目标车辆,且具有良好的鲁棒性。     

嵌入卡尔曼预测器的粒子滤波目标跟踪算法*

针对经典的粒子滤波视频目标跟踪算法进行粒子传播采用随机游走的方式,以及传统颜色直方图无法反映目标空间特征的问题,提出了一种改进的基于颜色的粒子滤波目标跟踪算法。该算法在统计目标二阶颜色直方图的基础上,获得粒子的观察概率密度函数,利用卡尔曼滤波确定粒子动态传播模型中的确定性漂移部分,使粒子状态估计值分布更精确地趋向目标的概率分布,大大提高了粒子的利用效率。实验表明,该改进算法的性能优于经典基于单一颜色特征的粒子滤波算法。     

基于CKF 的分布式滤波算法及其在目标跟踪中的应用

针对已有基于Sigma点信息滤波的分布式滤波算法,其性能易受参数影响而导致应用范围受限的问题,以容积卡尔曼滤波(CKF)为基础,利用信息滤波和平均一致性理论提出一种分布式CKF算法。该算法在保持分布式滤波优良特性(即可扩展性和对节点故障强鲁棒性)的同时,兼具CKF的高滤波精度和强稳定性。仿真结果表明了所提出算法的有效性,与分布式Unscented卡尔曼滤波(UKF)算法相比,该算法显著提高了目标跟踪的精度和稳定性。