一种新的基于粒子滤波的多模型跟踪算法

粒子滤波技术通过非参数化的蒙特卡罗模拟方法实现递推贝叶斯滤波,适用于非线性目标运动模型、非线性传感器测量模型和非高斯噪声的目标跟踪。但需已知目标和量测模型,而实际情况往往难以满足此条件。交互多模型算法(IMM)依据各模型对目标前一时刻状态估计的方差,确定各模型在当前时刻状态下存在的概率,利用各模型对目标状态估计的加权和,确定目标的状态。本文采用粒子滤波代替IMM算法中各模型的Kalman滤波,将粒子滤波与IMM的优点相结合。同时,采用UKF(UnscentedKalmanFilter)产生粒子,由于考虑了当前量测,使得粒子的分布更加接近后验概率分布,用较少的粒子就可以逼近目标的真实状态。仿真实验结果表明,本算法可用于标准IMM算法无法实现跟踪的复杂情形,而且使用的粒子数目仅是同类算法的二十分之一。     

Grip Control Using Biomimetic Tactile Sensing Systems

We present a proof-of-concept for controlling the grasp of an anthropomorphic mechatronic prosthetic hand by using a biomimetic tactile sensor, Bayesian inference, and simple algorithms for estimation and control. The sensor takes advantage of its compliant mechanics to provide a triaxial force sensing end-effector for grasp control. By calculating normal and shear forces at the fingertip, the prosthetic hand is able to maintain perturbed objects within the force cone to prevent slip. A Kalman filter is used as a noise-robust method to calculate tangential forces. Biologically inspired algorithms and heuristics are presented that can be implemented online to support rapid, reflexive adjustments of grip.      

基于SMSS-UKF的机载单站无源定位算法

基于机载单站无源定位中常用的扩展卡尔曼滤波具有近似线性方程的雅可比矩阵计算困难、不敏卡尔曼滤波(UKF)具有计算量较大的问题,提出了基于改进的变尺度最小斜度不敏卡尔曼滤波(SMSS-UKF)的机载单站无源定位算法,该方法采用最小斜度采样策略进行Sigma 点采样,减少Sigma点提高计算效率,利用变尺度不敏变换克服了采样点非局部效应问题。同时,针对大部分定位跟踪模型中状态方程为线性方程的特点,依据在线性状态方程情况下的贝叶斯理论,运用卡尔曼滤波状态预测的方法进行UKF的最优状态预测,使状态预测避免了不敏变换的数值近似误差和Sigma采样点计算的复杂性,提高了算法的运行效率和滤波精度。仿真实验的结果证明了SMSS-UKF滤波算法的有效性。     

Multiple camera tracking of interacting and occluded human motion

We propose a distributed, real-time computing platform for tracking multiple interacting persons in motion. To combat the negative effects of occlusion and articulated motion we use a multiview implementation, where each view is first independently processed on a dedicated processor. This monocular processing uses a predictor-corrector filter to weigh reprojections of three-dimensional (3-D) position estimates, obtained by the central processor, against observations of measurable image motion. The corrected state vectors from each view provide input observations to a Bayesian belief network, in the central processor, with a dynamic, multidimensional topology that varies as a function of scene content and feature confidence. The Bayesian net fuses independent observations from multiple cameras by iteratively resolving independency relationships and confidence levels within the graph, thereby producing the most likely vector of 3-D state estimates given the available data. To maintain temporal continuity, we follow the network with a layer of Kalman filtering that updates the 3-D state estimates. We demonstrate the efficacy of the proposed system using a multiview sequence of several people in motion. Our experiments suggest that, when compared with data fusion based on averaging, the proposed technique yields a noticeable improvement in tracking accuracy     

Intelligent Dynamic Radio Tracking in Indoor Wireless Local Area Networks

Indoor positioning is an enabling technology for delivery of location-based services in mobile computing environments. This paper proposes a positioning solution using received signal strength in indoor Wireless Local Area Networks. In this application, an explicit measurement equation and the corresponding noise statistics are unknown because of the complexity of the indoor propagation channel. To address these challenges, we introduce a new state-space Bayesian filter: the Nonparametric Information (NI) filter. This filter effectively tracks motion in situations where the Kalman filter and its variants are inapplicable, while maintaining a computational complexity comparable to that of the Kalman filter. To deal with the noisy nature of the indoor propagation environment, the NI filter is used in the design of an intelligent dynamic WLAN tracking system. The system anticipates future position values and adapts its sensing and estimation parameters accordingly. Our experimental results conducted on measurements from a real office environment indicate that the combination of the intelligent design and the NI filter results in significant improvements over the Kalman and particle filters.     

Generalized impedance control of robot for assembly tasks requiringcompliant manipulation

To perform assembly tasks requiring compliant manipulation, the robot must follow a motion trajectory and exert an appropriate force profile while making compliant contact with a dynamic environment. For this purpose, a generalized impedance in the task space consisting of a second-order function relating the motion errors and interaction force errors is introduced such that the contact force can be commanded and controlled. With generalized impedance control, the robot can behave with a desired dynamic characteristic when it interacts with the environment. To ensure the success of the assembly, a strategy during task planning which takes into consideration the interrelation between motion and force trajectories as well as contact compliance is introduced. The generalized impedance control method is applied to the prismatic joint of a selective compliance assembly robot arm (SCARA) robot for inserting a printed circuit board (PCB) into an edge connector socket. Depending on the progress of the parts joining operation, various amount of interaction forces are generated which have to be accommodated. It is demonstrated that an assembly strategy which consists of a sequence of carefully planned target impedance can enable the task to be executed in a desirable manner. The effectiveness of this approach is illustrated through experiments by comparing the results with those obtained using a well-established position control scheme as well as the original impedance control method     

基于测量协方差离散Kalman滤波估计算法的视频跟踪

视频中人体跟踪存在复杂性,尤其是对复杂背景下的人体上、下肢区域进行识别与跟踪时,传统算法存在一些问题。本文在传统Kalman滤波跟踪算法基础上,提出一种基于可变测量协方差的离散Kalman滤波人体识别算法。通过初始化测量协方差,用递归的方法从新获取的观测数据中计算出新的测量协方差估计量,通过离散Kalman滤波器进行跟踪。在实际的视频图像中,表现出良好的跟踪效果,并且对上肢、下肢及整个人体的区分以及部位跟踪方面都有很好的表现。相对于传统的Kalman滤波算法,本算法没有丢失跟踪目标的现象,跟踪速度适中,与人体行进速度保持一致,基本为1.5 m/s,特别适用于对视频中的人体行为进行跟踪及分析处理。     

渐进扩展卡尔曼滤波器

渐进贝叶斯方法将贝叶斯更新步骤等效为伪时间上的连续演化过程,以实现对状态的后验估计.本文基于渐进贝叶斯框架,导出一种新的高斯型非线性滤波算法.在线性高斯条件下推导了渐进贝叶斯方法的精确解;证明了对于由线性高斯解确定的动态系统,其均值和协方差矩阵满足的微分方程与常数状态估计的Kalman-Bucy滤波器是一致的.对于非线性系统,利用一阶Taylor展开推导了近似解表达式,进而导出渐进扩展卡尔曼滤波器.仿真算例表明新滤波器性能较扩展卡尔曼滤波器有大幅提高,且避免了窄形似然函数带来的滤波性能恶化问题.     

基于卡尔曼滤波器的焊缝识别算法

为了提高工业现场焊缝跟踪的准确性和抗干扰能力,提出了一种基于约束卡尔曼滤波器的焊缝识别算法。采用结构光测量系统获取焊缝连续图像,然后利用形态学处理定位焊缝图像角点,并通过高斯拟合法提取激光条纹中心点,将角点和中心点的位置和速度作为卡尔曼滤波器的状态变量,建立卡尔曼滤波器的状态方程和测量方程,再由角点和中心点的位置关系构建等式约束方程,通过约束方程对卡尔曼滤波后的结果进行修正。结果表明该算法能有效实现焊缝跟踪,提高了焊缝识别的精度和抗干扰能力。     

Practical Kalman Filter Software Performance Testing & Validation

Several tests are described which can be used for any Kalman-type filter/smoother computer program. These tests are demonstrated by a case history on a large dimensional Kalman filter/smoother program which implements a 34-state inertial navigation system dynamic error model. The execution of a large dimensional Kalman filter/smoother (KFS) on real measurement data does not represent a software test of the KFS since the right answer (the correct underlying state vector) is unknown; only ``reasonableness checks' are actually possible. Simulated test data were used to exercise the KFS program in a Monte Carlo sense and its outputs evaluated using heuristic plot comparisons as well as rigorous statistical tests. Direct tests on the accuracy of the transition matrix, discrete process noise matrix, and covariance matrix calculations have been derived and demonstrated. Methods for testing properties of the Kalman filter innovations sequence are also covered. The approach and required auxiliary software that generates the test data can be employed to perform suboptimal modeling sensitivity studies and for evaluating analysis methods that depend on KFS estimates.     

A Nonlinear Filtering Approach for Robust Multi-GNSS RTK Positioning in Presence of Multipath and Ionospheric Delays

In this paper, we develop a new approach of precise positioning using three carrier phase multi-Global Navigation Satellite System (GNSS) measurements in presence of multipath and ionospheric delays. We propose a new nonlinear filter to estimate the user position as well as all the unknown parameters including the integer ambiguities and the ionospheric errors. First, we use a kernel representation of the conditional density and apply a local linearization which yields a Kalman-like correction enhancing the particle filter correction. This new particle Kalman filter approach, is designed to be efficient for the non-Gaussian state and nonlinear measurements model, reduces the number of needed particles, and reduces the risk of divergence. The proposed procedure for multifrequency ambiguity resolution is based on four steps: 1) at each epoch, we compute the float solution adaptively to the dynamic environment by minimizing the noise level and estimating the ionospheric errors using the proposed robust Bayesian particle Kalman filter (RobPKF); 2) we introduce a new carrier phase multipath indicator and use it to derive a related constraint to reject integers candidates that are affected by multipath errors; 3) we apply the LAMBDA method to search the integer ambiguities; and finally 4) validate the fixed solution using a statistical test. We show in this work that the efficient integration of multifrequency/multisystem carriers provides more redundancy in the measurements and better observability for multipath and ionospheric errors estimation for long-baseline RTK positioning. A major advantage of this method is that it is independent of frequencies choice and therefore can be applied for any multi-GNSS measurements (e.g., Global Positioning System (GPS), Galileo, and their combination). Real-time and postprocessing test results show the effectiveness of the developed overall real-time kinematic (RTK) software.     

基于改进Camshift算法的NAO机器人目标跟踪

针对Camshift算法应用于NAO机器人目标跟踪过程中,当目标受到相似颜色背景干扰或被物体遮挡时跟踪失败的问题,提出一种基于ORB特征检测和Kalman滤波多算法结合的目标跟踪方法。首先检测目标ORB特征点初始化搜索窗口,然后利用Kalman滤波作为目标运动状态的预测机制,以预测的位置初始化Camshift算法。利用Bhattacharyya距离判断跟踪窗口的收敛性,若受到背景干扰,则利用ORB算法对当前帧中的Kalman预测区域和目标模型进行特征点匹配,重新检测目标在视频帧中的位置。根据Kalman滤波预测目标被物体遮挡后可能的位置来更新预测器参数。实验结果表明,改进的算法能够在相似颜色背景干扰和目标遮挡的复杂环境下,连续稳定地跟踪运动目标。     

三状态样条Kalman滤波与目标机动检测

基于样条方法和Kalman滤波理论，给出了一种新的三状态样条Kalman滤波方法及对目标进行机动检测的方法。该滤波方法用样条函数建立目标运动模型，应用Kalman滤波进行状态估计；该机动检测方法通过检测滤波新息来对目标机动发生或消除进行判断。由于考虑了加速度的实时估计，该滤波方法尤其适用于对弹道式目标的机动跟踪。仿真计算与实测数据计算表明：该滤波具有较高的精度；该机动检测方法具有较高的检测效率。     

Bayesian techniques for known channel deconvolution

This paper introduces a new family of deconvolution filters for digital communications subject to severe intersymbol interference. These fixed lag smoothing filters for known channel demodulation are called Bayesian filters. Bayesian filters are derived using a new approach to suboptimal recursive minimum mean square error estimation for non-Gaussian processes. The family of Bayesian filters interpolates between the optimum fixed lag linear filter (i.e., the Kalman filter) and the optimum fixed lag symbol-by-symbol demodulator in both performance and complexity. The complexity of the Bayesian filter is exponential in a parameter, typically chosen smaller than the channel length and the filter lag. Hence, the Bayesian filter decouples the channel length and the filter lag from the exponential complexity in these parameters found in many other high performance demodulation algorithms. Simulations characterize the performance and compare the Bayesian filter to both optimal and reduced complexity demodulation algorithms     

Joint Particle Filter and UKF Position Tracking in Severe Non-Line-of-Sight Situations

The performance of localization techniques in a wireless communication system is severely impaired by biases induced in the range and angle measures because of the non-line-of-sight (NLOS) situation, caused by obstacles in the transmitted signal path. However, the knowledge of the line-of-sight (LOS) or NLOS situation for each measure can improve the final accuracy. This paper studies the localization of mobile terminals (MT) based on a Bayesian model for the LOS-NLOS evolution. This Bayesian model does not require having a minimum number of LOS measures at each acquisition. A tracking strategy based on a particle filter (PF) and an unscented Kalman filter (UKF) is used both to estimate the LOS-NLOS situation and the MT kinetic variables (position and speed). The approach shows a remarkable reduction in positioning error and a high degree of scalability in terms of performance versus complexity.     

Bayesian Fusion for Indoor Positioning Using Bluetooth Fingerprints

This paper studies the use of received signal strength indicators (RSSI) applied to fingerprinting method in a Bluetooth network for indoor positioning. A Bayesian fusion (BF) method is proposed to combine the statistical information from the RSSI measurements and the prior information from a motion model. Indoor field tests are carried out to verify the effectiveness of the method. Test results show that the proposed BF algorithm achieves a horizontal positioning accuracy of about 4.7 m on the average, which is about 6 and 7 % improvement when compared with Bayesian static estimation and a point Kalman filter method, respectively.     

Nonlinear Bayesian Filtering Using the Unscented Linear Fractional Transformation Model

For nonlinear state space model involving random variables with arbitrary probability distributions, the state estimation given a sequence of observations is based on an appropriate criterion such as the minimum mean square error (MMSE). This leads to linear approximation in the state space of the extended Kalman filter (EKF) and the unscented Kalman filter (UKF), which work reasonably well only for mildly nonlinear systems. We propose a Bayesian filtering technique based on the MMSE criterion in the framework of the virtual linear fractional transformation (LFT) model, which is characterized by a linear part and a simple nonlinear structure in the feedback loop. LFT is an exact representation for any differentiable nonlinear mapping, so the virtual LFT model is amenable to a wide range of nonlinear systems. Simulation results demonstrate that the proposed filtering technique gives better approximation and tracking performance than standard methods like the UKF. Furthermore, for highly nonlinear systems where UKF diverges, the LFT model estimates the conditional mean with reasonable accuracy.      

Bayesian methods for multiaspect target tracking in image sequences

In this paper, we introduce new algorithms for automatic tracking of multiaspect targets in cluttered image sequences. We depart from the conventional correlation filter/Kalman filter association approach to target tracking and propose instead a nonlinear Bayesian methodology that enables direct tracking from the image sequence incorporating the statistical models for the background clutter, target motion, and target aspect change. Proposed algorithms include 1) a batch hidden Markov model (HMM) smoother and a sequential HMM filter for joint multiframe target detection and tracking and 2) two mixed-state sequential importance sampling trackers based on the sampling/importance resampling (SIR) and the auxiliary particle filtering (APF) techniques. Performance studies show that the proposed algorithms outperform the association of a bank of template correlators and a Kalman filter in adverse scenarios of low target-to-clutter ratio and uncertainty in the true target aspect.     

单雷达直线航迹线目标状态分量合成估计方法

通过分析传统卡尔曼滤波方法在复杂数据环境应用中遇到的问题,提出了基于单雷达加权直线航迹线参数估计模型的目标运动状态分量合成估计方法。该方法基于复杂数据环境,无须获取系统噪声和观测噪声等先验知识,充分利用目标处于匀速直线运动状态这一特点,分别对当前有限测量点的X、Y 分量进行相对于的测量时刻的垂直距离加权迭代估计,确定目标状态估计参数。通过对比试验,验证了文中所提的方法比传统卡尔曼方法具有更优的目标状态估计效果、测量误差平滑和野值抑制能力,能有效提高观测样本较少时目标状态参数估计的准确性。     

基于SUKF与SIFT特征的红外目标跟踪算法研究

针对复杂红外背景下单一跟踪算法难以准确定位运动目标的问题,提出了基于尺度无迹卡尔曼滤波(SUKF,scale unscented Kalman filter)与尺度不变特征变换(SIFT,scale invariant featuretransform)相结合的红外运动目标跟踪方法。首先,通过SUKF算法对状态空间进行滤波估计,确定运动目标的初步位置,并以此建立局部SIFT特征检测域。其次,SIFT算法在该局部检测域内对运动目标进行特征提取与匹配,最终实现对目标的准确定位;同时,利用定位结果更新并校正SUKF的状态模型。实验结果表明,本文提出的基于SUKF-SIFT的跟踪策略与相关算法相比,体现出较好的跟踪效果与实时性能。