Efficient multisensor fusion using multidimensional dataassociation

We present the development of a multisensor fusion algorithm using multidimensional data association for multitarget tracking. The work is motivated by a large scale surveillance problem, where observations from multiple asynchronous sensors with time-varying sampling intervals (electronically scanned array (ESA) radars) are used for centralized fusion. The combination of multisensor fusion with multidimensional assignment is done so as to maximize the “time-depth” in addition to “sensor-width” for the number S of lists handled by the assignment algorithm. The standard procedure, which associates measurements from the most recently arrived S-1 frames to established tracks, can have, in the case of S sensors, a time-depth of zero. A new technique, which guarantees maximum effectiveness for an S-dimensional data association (S⩾3), i.e., maximum time-depth (S-1) for each sensor without sacrificing the fusion across sensors, is presented. Using a sliding window technique (of length S), the estimates are updated after each frame of measurements. The algorithm provides a systematic approach to automatic track formation, maintenance, and termination for multitarget tracking using multisensor fusion with multidimensional assignment for data association. Estimation results are presented for simulated data for a large scale air-to-ground target tracking problem     

Tracking of Spawning Targets with Multiple Finite Resolution Sensors

In this paper the problem of tracking multiple spawning targets with multiple finite-resolution sensors is considered and a new algorithm for measurement-to-track association with possibly unresolved measurements is presented. The goal is to initialize new tracks of spawned targets before they are resolved from the mother platform so that one has the ability to carry out early discrimination when they become resolved. The multiple scan data association problem is first formulated as a multidimensional assignment problem with explicit new constraints for the unresolved measurements. Then the top M hypotheses tracking (TMHT) is presented where the state estimates and their covariances are modified based on the M best hypotheses through the assignment solutions. A modification to the assignment problem is developed that leads to a linear programming (LP) where the optimal solution can be a noninteger in [0,1]. The fractional optimal solution is interpreted as (pseudo) probabilities over the N - 1 frame sliding window. The best hard (binary) decision assignment solution and the M best via TMHT are compared with the soft decision solution for 2-D tracking scenarios with various sensor configurations. Based on the simulation results, the soft assignment approach has better track maintenance capability than the single best hard assignment and a performance nearly as good as the TMHT. Its computational load is slightly higher than the single best hard assignment but much lighter than TMHT.     

Dynamically adaptable m-best 2-D assignment algorithm andmultilevel parallelization

In recent years, there has been considerable interest within the tracking community in an approach to data association based on the m-best two-dimensional (2D) assignment algorithm. Much of the interest has been spurred by its ability to provide various efficient data association solutions, including joint probabilistic data association (JPDA) and multiple hypothesis tracking (MHT). The focus of this work is to describe several recent improvements to the m-best 2D assignment algorithm. One improvement is to utilize a nonintrusive 2D assignment algorithm switching mechanism, based on a problem sparsity threshold. Dynamic switching between two different 2D assignment algorithms, highly suited for sparse and dense problems, respectively, enables more efficient solutions to the numerous 2D assignment problems generated in the m-best 2D assignment framework. Another improvement is to utilize a multilevel parallelization enabling many independent and highly parallelizable tasks to be executed concurrently, including 1) solving the multiple 2D assignment problems via a parallelization of the m-best partitioning task, and 2) calculating the numerous gating tests, state estimates, covariance calculations, and likelihood function evaluations (used as cost coefficients in the 2D assignment problem) via a parallelization of the data association interface task. Using both simulated data and an air traffic surveillance (ATS) problem based on data from two Federal Aviation Administration (FAA) air traffic control radars, we demonstrate that efficient solutions to the data association problem are obtainable using our improvements in the m-best 2D assignment algorithm     

Suboptimal joint probabilistic data association

A significant problem in multiple target tracking is the hit-to-track data association. A hit is a received signal from a target or background clutter which provides positional information. If an incorrect hit is associated with a track, that track could diverge and terminate. Prior methods for this data association problem include various optimal and suboptimal two-dimensional assignment algorithms which make hit-to-track associations. Another method is to assign a weight for the reasonable hits and use a weighted centroid of those hits to update the track. The method of weighting the hits is known as joint probabilistic data association (JPDA). The authors review the JPDA approach and a simple ad hoc approximation and then introduce a new suboptimal JPDA algorithm. Examples which compare an optimal two-dimensional assignment algorithm with the ad hoc and the new suboptimal JPDA formulation are given     

Multitarget tracking in clutter: fast algorithms for dataassociation

Three fast algorithms have been developed to solve the problem of data association in multitarget tracking in clutter. In the first algorithm, the problem of data association is identified as an exhaustive search problem in general. Subsequently, a mathematical model is proposed for the problem of data association in the joint probabilistic data association filter (JPDAF). Based on the model, a depth-first search (DFS) approach is developed for the fast generation of data association hypotheses and the computation of the conditional probabilities of the hypotheses in the JPDAF. When the density of targets is moderate, a second algorithm is developed to directly compute a posteriori probabilities in the JPDAF without generating the data association hypotheses. In the third algorithm, the effect of interference due to closely spaced targets is simplified. An approach to approximately compute the a posteriori probabilities in the JPDAF is developed. The computational complexity of the algorithms is analyzed in the worst case, as well as in the average case     

An interior point linear programming approach to two-scan dataassociation

We present an efficient two-scan data association method (TSDA) based on an interior point linear programming (LP) approach. In this approach, the TSDA problem is first formulated as a 3-dimensional assignment problem, and then relaxed to a linear program; the latter is subsequently solved by the highly efficient homogeneous, self-dual interior point LP algorithm. When the LP algorithm generates a fractional optimal solution, we use a technique similar to the joint probabilistic data association method (JPDA) to compute a weighted average of the resulting fractional assignments, and use it to update the states of the existing tracks generated by Kalman filters. Unlike the traditional single scan JPDA method, our TSDA method provides an explicit mechanism for track initiation. Extensive computer simulations have demonstrated that the new TSDA method is not only far more efficient in terms of low computational complexity, but also considerably more accurate than the existing single-scan JPDA method     

Mixture Reduction Algorithms for Point and Extended Object Tracking in Clutter

The Bayesian solution to the problem of tracking a target with measurement association uncertainty gives rise to mixture distributions, which are composed of an ever increasing number of components. To produce a practical tracking filter, the growth of components must be controlled by approximating the mixture distribution. Two mixture reduction schemes (a joining algorithm and a clustering algorithm) have been derived for this purpose. If significant well spaced mixture components are present, these techniques can provide a useful improvement over the probabilistic data association filter (PDAF) approach, which reduces the mixture to a single Gaussian component at each time step. For the standard problem of tracking a point target in uniform random clutter, a Monte Carlo simulation study has been employed to identify the region of the problem parameter space where significant performance improvement is obtained over the PDAF. In the second part of this paper, the formal Bayesian filter is derived for an extended target consisting of an array of measurement sources with association uncertainty. A practical multiple hypothesis filter is implemented using mixture reduction and simulation results are presented.     

New assignment-based data association for tracking move-stop-move targets

We present a new assignment-based algorithm for data association in tracking ground targets employing evasive move-stop-move maneuvers using ground moving target indicator (GMTI) reports obtained from an airborne sensor. To avoid detection by the GMTI sensor, the targets deliberately stop for some time before moving again. The sensor does not detect a target when the latter's radial velocity (along the line-of-sight from the sensor) falls below a certain minimum detectable velocity (MDV). Even in the absence of move-stop-move maneuvers, the detection has a less-than-unity probability (P/sub D/<1) due to obscuration and thresholding. Then, it is of interest, when a target is not detected, to develop a systematic technique that can distinguish between lack of detection due to P/sub D/<1 and lack of detection due to a stop (or a near stop). Previously, this problem was solved using a variable structure interacting multiple model (VS-IMM) estimator with a stopped target model (VS-IMM-ST) without explicitly addressing data association. We develop a novel "two-dummy" assignment approach for move-stop-move targets that considers both the problem of data association as well as filtering. Typically, in assignment-based data association a "dummy" measurement is used to denote the nondetection event. The use of the standard single-dummy assignment, which does not handle move-stop-move motion explicitly, can result in broken tracks. The new algorithm proposed here handles the evasive move-stop-move motion by introducing a second dummy measurement to represent nondetection due to the MDV. We also present a likelihood-ratio-based track deletion scheme for move-stop-move targets. Using this two-dummy data association algorithm, the track corresponding to a move-stop-move target is kept "alive' during missed detections both due to MDV and due to P/sub D/<1. In addition, one can obtain reductions in both rms estimation errors as well as the total number of track breakages.     

Extracting multiple frequencies from phase-only data

The problem of extracting multiple frequencies from phase-only data is addressed. Multiple frequency estimation is accomplished by reconstructing the Fourier transform of the complex-valued time signal and then finding peaks in the frequency domain. We present a set of conditions under which a discrete-time complex sequence can be completely specified by its phase-only information. Two candidate multiple frequency estimation schemes are introduced, one based on a closed-form least-squares inverse, the other an iterative reconstruction algorithm. The uniqueness of the closed-form solution and the convergence of the iterative scheme have been proven under certain conditions. Several examples are given, including the case where the phase is quantized as would happen in an analog-to-digital (A/D) converter. Extensions to the multidimensional case, and to the case of real-part only reconstruction are straightforward.     

Tracking multiple objects with particle filtering

We address the problem of multitarget tracking (MTT) encountered in many situations in signal or image processing. We consider stochastic dynamic systems detected by observation processes. The difficulty lies in the fact that the estimation of the states requires the assignment of the observations to the multiple targets. We propose an extension of the classical particle filter where the stochastic vector of assignment is estimated by a Gibbs sampler. This algorithm is used to estimate the trajectories of multiple targets from their noisy bearings, thus showing its ability to solve the data association problem. Moreover this algorithm is easily extended to multireceiver observations where the receivers can produce measurements of various nature with different frequencies.     

Multiassignment for tracking a large number of overlapping objects[and application to fibroblast cells]

In this paper we present a new technique for data association using multiassignment for tracking a large number of closely spaced (and overlapping) objects. The algorithm is illustrated on a biomedical problem, namely the tracking of a group of fibroblast (tissue) cells from an image sequence, which motivated this work. Because of their proximity to one another and due to the difficulties in segmenting the images accurately from a poor-quality image sequence, the cells are effectively closely spaced objects (CSOs). The algorithm presents a novel dichotomous, iterated approach to multiassignment using successive one-to-one assignments of decreasing size with modified costs. The cost functions, which are adjusted depending on the “depth” of the current assignment level and on the tracking results, are derived. The resulting assignments are used to form, maintain and terminate tracks with a modified version of the probabilistic data association (PDA) filter, which can handle the contention for a single measurement among multiple tracks in addition to the association of multiple measurements to a single track. Estimation results are given and compared with those of the standard 2D one-to-one assignment algorithm. It is shown that iterated multiassignment results in superior measurement-to-track association. The algorithms presented can be used for other general tracking problems, including dense air traffic surveillance and control     

Data association for deghosting in Y-shaped passive linear array sonars

This paper deals with data association using three sets of passive linear array sonars (PLAS) geometrically positioned in a Y-shaped configuration, fixed in an underwater environment. The data association problem is directly transformed into a 3D assignment, which is known to be NP hard. For generic passive sensors, it can be solved using conventional algorithms, while in PLAS, it becomes a formidable task due to the presence of bearing ambiguity. Thus, the central issue of the problem in PLAS is how to eliminate the bearing ambiguity without increasing tracking error. To solve this problem, the 3D assignment algorithm used the likelihood value of only those observed bearing measurements is modified by incorporating frequency information in consecutive time-aligned scans. The region of possible ghost targets is first established by the geometrical relation of PLAS with respect to target. The ghost targets are then confirmed and eliminated by generating multiple observations in consecutive scans. Representative simulations demonstrate the effectiveness of the proposed approach.     

Ground target tracking with variable structure IMM estimator

In this paper we present the design of a Variable Structure Interacting Multiple Model (VS-IMM) estimator for tracking groups of ground targets on constrained paths using Moving Target Indicator (MTI) reports obtained from an airborne sensor. The targets are moving along a highway, with varying obscuration due to changing terrain conditions. In addition, the roads can branch, merge or cross-the scenario represents target convoys along a realistic road network with junctions, changing terrains, etc. Some of the targets may also move in an open field. This constrained motion estimation problem is handled using an IMM estimator with varying mode sets depending on the topography, The number of models in the IMM estimator, their types and their parameters are modified adaptively, in real-time, based on the estimated position of the target and the corresponding road/visibility conditions. This topography-based variable structure mechanism eliminates the need for carrying all the possible models throughout the entire tracking period as in the standard IMM estimator, significantly improving performance and reducing computational load. Data association is handled using an assignment algorithm. The estimator is designed to handle a very large number of ground targets simultaneously. A simulated scenario consisting of over one hundred targets is used to illustrate the selection of design parameters and the operation of the tracker. Performance measures are presented to contrast the benefits of the VS-IMM estimator over the Kalman filter and the standard IMM estimator, The VS-IMM estimator is then combined with multidimensional assignment to gain “time-depth.” The additional benefit of using higher dimensional assignment algorithms for data association is also evaluated     

Tracking with classification-aided multiframe data association

In most conventional tracking systems, only the target kinematic information from, for example, a radar or sonar or an electro-optical sensor, is used in measurement-to-track association. Target class information, which is typically used in postprocessing, can also be used to improve data association to give better tracking accuracy. The use of target class information in data association can improve discrimination by yielding purer tracks and preserving their continuity. In this paper, we present the simultaneous use of target classification information and target kinematic information for target tracking. The approach presented integrates target class information into the data association process using the 2-D (one track list and one measurement list) as well as multiframe (one track list and multiple measurement lists) assignments. The multiframe association likelihood is developed to include the classification results based on the "confusion matrix" that specifies the accuracy of the target classifier. The objective is to improve association results using class information when the kinematic likelihoods are similar for different targets, i.e., there is ambiguity in using kinematic information alone. Performance comparisons with and without the use of class information in data association are presented on a ground target tracking problem. Simulation results quantify the benefits of classification-aided data association for improved target tracking, especially in the presence of association uncertainty in the kinematic measurements. Also, the benefit of 5-D (or multiframe) association versus 2-D association is investigated for different quality classifiers. The main contribution of this paper is the development of the methodology to incorporate exactly the classification information into multidimensional (multiframe) association.     

一种基于模糊遗传算法的多传感器多目标跟踪数据关联算法

基于模糊遗传算法发展了一种新的数据关联算法。数据关联的静态部分靠一个模糊遗传算法来得出量测组合序列和S-D分配的m个最优解。在数据关联的动态部分，将得到的S-D分配的m个最优解在一个基于多种群模糊遗传算法的动态2D分配算法中依靠一个卡尔曼滤波估计器估计出移动目标各个时刻的状态。这一基于分配的数据关联算法的仿真试验内容为被动式传感器的航迹形成和维持的问题。仿真试验的结果表明该算法在多传感器多目标跟踪中应用的可行性。另外，对算法发展和实时性问题进行了简单讨论。     

Track labeling and PHD filter for multitarget tracking

Multiple target tracking requires data association that operates in conjunction with filtering. When multiple targets are closely spaced, the conventional approaches (as, e.g., MHT/assignment) may not give satisfactory results. This is mainly because of the difficulty in deciding what the number of targets is. Recently, the probability hypothesis density (PHD) filter has been proposed and particle filtering techniques have been developed to implement the PHD filter. In the particle PHD filter, the track labeling problem is not considered, i.e., the PHD is obtained only for a frame at a time, and it is very difficult to perform the multipeak extraction, particularly in high clutter environments. A track labeling method combined with the PHD approach, as well as considering the finite resolution, is proposed here for multitarget tracking, i.e., we keep a separate tracker for each target, use the PHD in the resolution cell to get the estimated number and locations of the targets at each time step, and then perform the track labeling ("peak-to-track" association), whose results can provide information for PHD peak extraction at the next time step. Besides, by keeping a separate tracker for each target, our approach provides more information than the standard particle PHD filter. For example, in group target tracking, if we are interested in the motion of a specific target, we can track this target, which is not possible for the standard particle PHD filter, since the standard particle PHD filter does not keep track labels. Using our approach, multitarget tracking can be performed with automatic track initiation, maintenance, spawning, merging, and termination     

Behavior pattern mining based on spatiotemporal trajectory multidimensional information fusion

Trajectory data mining is widely used in military and civil applications, such as early warning and surveillance system, intelligent traffic system and so on. Through trajectory similarity measurement and clustering, target behavior patterns can be found from massive spatiotemporal trajectory data. In order to mine frequent behaviors of targets from complex historical trajectory data, a behavior pattern mining algorithm based on spatiotemporal trajectory multidimensional information fusion is proposed in this paper. Firstly, spatial–temporal Hausdorff distance is proposed to measure multidimensional information differences of spatiotemporal trajectories, which can distinguish the behaviors with similar location but different course and velocity. On this basis, by combining the idea of k-nearest neighbor and density peak clustering, a new trajectory clustering algorithm is proposed to mine behavior patterns from trajectory data with uneven density distribution. Finally, we implement the proposed algorithm in simulated and radar measured trajectory data respectively. The experimental results show that the proposed algorithm can mine target behavior patterns from different complex application scenarios more quickly and accurately compared to the existing methods, which has a good application prospect in intelligent monitoring tasks.     

密集杂波环境下的数据关联快速算法

基于联合概率数据互联（JPDA）的思想,提出了一种新的数据关联快速算法（Fast Al-gorithm for Data Association,简称FAFDA算法）.该方法不需象在最优JPDA算法中那样生成所有可能的联合互联假设,因而具有计算量小,易于工程实现的特点。仿真结果表明,与最优JPDA算法相比,FAFDA算法的跟踪性能令人满意,并且在密集杂波环境下可实时、有效地跟踪100批次以上的目标。     

Multiple scan joint probabilistic data association

This paper presents a multiple scan or n-scan-back joint probabilistic data association (JPDA) algorithm which addresses the problem of measurement-to-track data association in a multiple target and clutter environment. The standard single scan JPDA algorithm updates a track with weighted sum of the measurements which could have reasonably originated from the target in track. The only information the standard JPDA algorithm uses is the measurements on the present scan and the state vectors and covariance matrices of the present targets. The n-scan-back algorithm presented here uses multiple scans of measurements along with the present target information to produce better weights for data association. The standard JPDA algorithm can utilize a formidable amount of processing power and the n-scan-back version only worsens the problem. Therefore, along with the algorithm presentation, implementations which make this algorithm practical are discussed and referenced. An example is also shown for a few n-scan-back window lengths     

基于聚类的星座卫星可用性分选方法

卫星可用状况会随时间发生变化,为了实现对卫星工作情况的监控,应经常对目标卫星进行分析。通过实际侦收卫星信号确定卫星可用状况是最可靠的方法,但这种方法工作量大、周期长、效率低、实时性差。为了解决这一问题,本文提出一种运用聚类算法的卫星可用状况快速分选方法。该方法利用从互联网下载的最新轨道参数,根据卫星轨道特性方面的差别进行自动分选,具有速度快、人工干预少的优点。将这种方法用于铱星星座分选,分选正确率可达80.44%。