Tracking In Clutter Research Articles

Smoothing joint integrated probabilistic data association

The authors extend the smoothing integrated probabilistic data association algorithm to multi-target tracking in clutter, or alternatively, use smoothing to improve the joint integrated probabilistic data association (JIPDA) algorithm. The predictions of forward and backward JIPDA are fused to form the smoothing prediction, which is used for smoothing multi-target data association.

Efficient calculation of information reduction factor for tracking in clutter

A reformulation of the information reduction factor (IRF) for tracking in the presence of measurement uncertainty is presented. A simple integral representation is employed to derive an expression for the IRF that can be calculated without approximation in terms of one-dimensional integrals. This significantly reduces the computational requirements of the calculation and is relevant to the offline calculation of tracking performance and the selection of detection thresholds to optimise tracking accuracy. A comparison with Monte Carlo integration shows that the new technique is approximately 380 times faster.

LMMSE Filtering in Feedback Systems With White Random Modes: Application to Tracking in Clutter

A generalized state space representation of dynamical systems with random modes switching according to a white random process is presented. The new formulation includes a term, in the dynamics equation, that depends on the most recent linear minimum mean squared error (LMMSE) estimate of the state. This can model the behavior of a feedback control system featuring a state estimator. The measurement equation is allowed to depend on the previous LMMSE estimate of the state, which can represent the fact that measurements are obtained from a validation window centered about the predicted measurement and not from the entire surveillance region. The LMMSE filter is derived for the considered problem. The approach is demonstrated in the context of target tracking in clutter and is shown to be competitive with several popular nonlinear methods.

Linear multitarget finite resolution tracking in clutter

A majority of target tracking algorithms are derived assuming the infinite resolution sensor(s), where each measurement has only one source. This assumption is an approximation as the sensors (e.g., radars and sonars) have finite resolution, where close measurements from multiple sources may merge into one. This disparity between assumptions and reality may significantly degrade target tracking performance. This paper proposes a multitarget tracking algorithm that assumes finite resolution sensor(s). A finite resolution variant of linear multitarget tracking is proposed to limit the computational complexity. The proposed algorithm provides both the trajectory estimate and the track quality measure enabling automatic track initiation and termination (false track discrimination) for a complete target tracking solution in nonhomogeneous clutter. A simulation study vindicates the approach in a difficult situation with 10 crossing targets in clutter that may merge measurements.

다중표적 추적필터와 자료연관 기법동향

This paper is to survey and put in perspective the working methods of multi-target tracking in clutter. This paper includes theories and practices for data association and related filter structures and is motivated by increasing interest in the area of target tracking, security, surveillance, and multi-sensor data fusion. It is hoped that it will be useful in view of taking into consideration a full understanding of existing techniques before using them in practice.

A Hybrid Multi-sensor Multi-target Tracking Scheme with MLE and ANFIS

The Joint Probabilistic Data Association (JPDA) solves single sensor multi-target tracking in clutter, but it cannot be used directly in multi-sensor multi-target tracking (MMT) and has high computational complexity with the number of targets and the number of returns. This paper presents a hybrid method to implement MMT by combing Maximum Likelihood Estimation (MLE) with Adaptive Neuro-Fuzzy Inference System (ANFIS). The MLE is applied to classify the same source observations at one time into the same set, then the cheap JPDA(CJPDA) approach is used to calculate the data association probability, and ANFIS is used to realize the MMT. The computer simulations indicate that this scheme achieves MMT perfectly with higher precision and easy realization.

Gaussian mixture reduction based on fuzzy ART for extended target tracking

This paper presents a global Gaussian mixture (GM) reduction algorithm via clustering for extended target tracking in clutter. The proposed global clustering algorithm is obtained by combining a fuzzy Adaptive Resonance Theory (ART) neural network architecture with the weighted Kullback–Leibler (KL) difference which describes discrimination of one component from another. Therefore, we call the proposed algorithm as ART-KL clustering (ART-KL-C) in the paper. The weighted KL difference is used as a category choice function of ART-KL-C, derived by considering both the KL divergence between two components and their weights. The performance of ART-KL-C is evaluated by the normalized integrated squared distance (NISD) measure, which describes the deviation between the original and reduced GM. The proposed algorithm is tested on both one-dimensional and four-dimensional simulation examples, and the results show that the proposed algorithm can more accurately approximate the original mixture and is useful in extended target tracking.

Single Maneuvering Target Tracking in Clutter Based on Multiple Model Algorithm with Gaussian Mixture Reduction

The measurement origin uncertainty and target (dynamic or/and measurement) model uncertainty are twofundamental problems in maneuvering target tracking in clutter. The multiple hypothesis tracker (MHT) and multiplemodel (MM) algorithm are two well-known methods dealing with these two problems, respectively. In this work, weaddress the problem of single maneuvering target tracking in clutter by combing MHT and MM based on the Gaussianmixture reduction (GMR). Different ways of combinations of MHT and MM for this purpose were available in previousstudies, but in heuristic manners. The GMR is adopted because it provides a theoretically appealing way to reducethe exponentially increasing numbers of measurement association possibilities and target model trajectories. Thesuperior performance of our method, comparing with the existing IMM+PDA and IMM+MHT algorithms, isdemonstrated by the results of Monte Carlo simulation.

Smoothing Multi-Scan Target Tracking in Clutter

This paper presents a fixed interval smoothing multi-scan algorithm for target tracking in clutter. Both the probability of target existence and the target trajectory probability density function are calculated using all available measurements. This improves both the false track discrimination and the target trajectory estimate. The fixed interval smoothing fuses the forward and the backward multi-scan predictions, to obtain the smoothing predictions and smoothing innovations. Both trajectory estimates and the data association probabilities are calculated using the smoothing innovations. An overlapping batch procedure is described which limits the smoothing delay.

An advanced Data Mining Strategy for Multi-Target Tracking in Adverse Environments

For real-time tracking, simple and accurate data mining is required. The detection process produces a number of clutter measurements in addition to the target measurements and the data mining are important to stochastically distinguish between the target and the clutter measurements. In this paper we propose a computationally efficient data mining algorithm which is combined with a particle filter for a computationally efficient solution for nonlinear target tracking in clutter. True tracks are confirmed, and false tracks are terminated in a manner similar to the Integrated PDA. Monte Carlo simulation studies of multi-target tracking show that the proposed algorithm is capable of good performance in adverse environments

A Parallel PDA Algorithm for Closely-Spaced Targets Tracking in Clutter

To reduce the computational complexity of JPDA and resolve closely-spaced targets efficiently, decompose JPDA into parallel PDA algorithms without regarding the correlation of targets. For a target, measurements are all clutter except the ones originated from itself. The PDA algorithm to track every single target with fixed covariance of states. When targets got closed within thick clutter, try to avoid coalescence and reduce complexity of association, maintaining the independence of targets through neglecting the assignment association model of measurement-to-track which used to happen once targets gates got overlapped. Simulations give arguments that parallel PDA has a better performance than JPDA in tracking closely-spaced targets.

Dynamic waveform selection for manoeuvering target tracking in clutter

In recent years, cognitive radar (CR) with waveform diversity has exhibited significant performance improvements over the traditional fixed waveform radar and become an area of vigorous research and development. This study presents a dynamic waveform selection algorithm to strive for tracking error minimisation for CR manoeuvering target tracking in clutter. Based on the concepts of resolution cell and measurement extraction cell, the statistical characteristics of radar measurements are discussed without dependence upon the Cramer-Rao lower bound of the measurement errors and the high signal-to-noise ratio assumption. A particle filter combined with probabilistic data association is used as a tracker. To quantify the utility of available waveforms, the predicted tracking mean-square error, because of its dependence on actual future measurements, is approximated efficiently via Gaussian fitting of the prior density of the target state and statistical linearisation of the measurement equation. Monte Carlo simulation results show that the proposed dynamic waveform selection algorithm can improve tracking performance considerably, especially in terms of track loss probability.

Offline Performance Prediction of PDAF With Bayesian Detection for Tracking in Clutter

Conventional detectors in tracking system generally work according to the Neyman-Pearson criterion. Recently, Bayesian detection has become an alternative for many trackers such as probabilistic data association filter (PDAF). However, a critical problem of the tracking system with Bayesian detection is to predict the tracking performance without simulations. As the Bayesian detection is introduced, clutters are nonuniformly distributed and the detection threshold varies with time, which increases the difficulty of the analysis. In this paper, an offline method is developed to predict the performance of the PDAF with Bayesian detection (PDAF-BD). In the approach, the information reduction factor (IRF) of the PDAF-BD is derived, describing the influence of measurement origin uncertainty. Unlike the IRF of PDAF, the IRF of PDAF-BD has analytical expression which is efficient in computation. On this basis, the offline recursion of the error covariance and the quantification of track loss are achieved. The experiments show that the nonsimulated result generated by the proposed algorithm is reasonably close to the simulated one.

Sensor Fusion and Target Tracking Using Evidential Data Association

The problem of multitarget tracking in clutter has been shown to be very challenging for both the measurement of track association and the estimation of the targets' state. Several approaches have been suggested to solve this problem. In this paper, we present a new method for data association in multitarget tracking in the framework of evidence theory. The representation and the fusion of the information in our method are based on the use of belief function in the sense of Dempster-Shafer theory of evidence. The proposal introduces a method of computing mass assignment using a normalized Mahalanobis distance. While the decision making process is based on the extension of the frame of hypotheses, the method has been tested for a nearly constant velocity target and compared with both the nearest neighbor filter and the joint probabilistic data associations filter in highly ambiguous cases using Monte Carlo simulations. The results demonstrate the feasibility of the proposal, and show improved performance compared to the aforementioned alternative commonly used methods.

Correlated Doppler‐assisted target tracking in clutter

In a surveillance situation, the origin of each measurement is uncertain. Each measurement may be a clutter measurement or a target detection. In addition to the position [range and angle(s)], measurements often include Doppler velocity. The Doppler velocity measurements are often (e.g. with the linearly modulated frequency transmitted signal) correlated with the range measurements. This study deals with data association (via measurement likelihoods) and estimation in this situation, with a special consideration of the Gaussian mixture approaches. A simulation study shows the benefit of this approach.

A NOVEL INTERACTING MULTIPLE MODEL PARTICLE FILTER FOR MANEUVERING TARGET TRACKING IN CLUTTER

In this paper, a novel interactive multiple model particle fllter (IMMPF) is developed after a Bayesian estimator for maneuvering target tracking in clutter is derived theoretically. In this new algorithm, base state estimation and modal state estimation are completely separated to control the number of particles in each maneuvering mode. Only continuous-valued particles are used to numerically implement the procedure of Bayesian base state estimation, whereas modal state is estimated analytically without dependence on the number of particles. Density mixing is performed by aggregation of the total particles and mixing associated weights. To prevent the exponentially growing number of particles with the time, a resampling step is included following the interaction step. Through MC simulations, the new IMMPF has been tested and shown to provide reliable performance improvements with difierent sample sizes and under various clutter conditions.

Non-linear automatic target tracking in clutter using dynamic Gaussian mixture

This study presents a complete algorithm for single target tracking in clutter, which addresses simultaneously: non-linear measurements; uncertain target detections; presence of random clutter measurements; and uncertain target existence. Proposed algorithm generalises the integrated track splitting (ITS) filter by extending the ITS functionality to highly non-linear measurements. The non-linear target tracking and estimation problems may also be solved by application of particle filters, albeit incurring a significant computational expense relative to proposed solution. In an environment without data association uncertainties proposed filter becomes a non-linear estimator.

Directional Fuzzy Data Association Filter

In this paper, a new multi-target tracking algorithm based on fuzzy logic for tracking in clutter is developed, it is called directional fuzzy data association (DFDA) filter. The new algorithm incorporates the directional information of the targets for data association with the Mahalanobis distance. Firstly, the directional information, called pseudo-direction, is defined; the method of how to calculate the pseudo-direction has been introduced. Then the state incorporating with the pseudo-direction is updated using the cubature Kalman filter (CKF). At last the fuzzy logic inference method is used for data association. Simulation results are used to evaluate the performance of this new algorithm comparing with the nearest neighbor standard filter (NNSF) and joint probability data association filter (JPDAF), the final results show that the proposed DFDA filter an efficient and effective approach for real application.

Class-Dependent Gating Algorithm in Data Association

Most conventional tracking gate algorithms only use the targets’ kinematic measurement information, which is typically resulted in great uncertainties of measurement-to-track association for multi-target tracking in clutter. The problem of constructing tracking gates using targets' class information is considered. The proposed algorithm integrates targets' identity information into the traditional tracking gating techniques. First, a class-dependent gate corresponding to each class of targets is developed. Second, the algorithm for constructing the class-dependent gate is given. Simulations are carried out to examine the proposed algorithm, where the simulation scenario shows that the measurement-to-track association using the class-dependent gating algorithm is significantly better than traditional method.

Bernoulli Particle Filter with Observer Control for Bearings-Only Tracking in Clutter

The context is autonomous bearings-only tracking of a single appearing/disappearing target in the presence of detection uncertainty (false and missed detections) with observer control. The optimal tracking method for this problem in the sequential Bayesian estimation framework is the Bernoulli filter. Observer control is based on previously acquired measurements and is formulated as a partially observable Markov decision process (POMDP) where future actions are ranked according to their associated reward. The paper develops a sequential Monte Carlo implementation of the Bernoulli filter and the reward based on an information theoretic criterion.