International Journal of Control Automation and Systems, Feb 1, 2008
An interacting multiple model (IMM) estimation algorithm based on the mixing of the predicted sta... more An interacting multiple model (IMM) estimation algorithm based on the mixing of the predicted state estimates is proposed in this paper for a right continuous jump-linear system model different from the left-continuous system model used to develop the existing IMM algorithm. The difference lies in the modeling of the mode switching time. Performance of the proposed algorithm is compared numerically with that of the existing IMM algorithm for noisy system identification. Based on the numerical analysis, the proposed algorithm is applied to target tracking with a large sampling period for performance comparison with the existing IMM.
A Gaussian Mixture (GM) target tracking solution is a natural consequence of the multi-target tra... more A Gaussian Mixture (GM) target tracking solution is a natural consequence of the multi-target tracking in clutter, given a linear target trajectory propagation and a linear target measurement equation. We examine and compare two prominent GM target trackers: the Multi Hypothesis Tracking (MHT) and the Integrated Track Splitting. Both incorporate the false track discrimination capabilities, enabling automatic target tracking in the presence of clutter measurements and missed detections.
Multistatic tracking involves using non-collocated transmitters and receivers to track the target... more Multistatic tracking involves using non-collocated transmitters and receivers to track the targets. In this paper we also assume no angle information. This paper proposes an approach using Gaussian Mixture representation of measurement uncertainty. An arbitrary number of receivers can be used using this approach. Here we limit ourselves to the track estimation issues, i.e. we assume no clutter measurement, and imperfect target detection at the receivers. Simulation results are presented which vindicate this approach.
IEEE Transactions on Aerospace and Electronic Systems, Apr 1, 2011
In a surveillance situation the origin of each measurement is uncertain. Each measurement may be ... more In a surveillance situation the origin of each measurement is uncertain. Each measurement may be a false (clutter) measurement, or it may be a target detection. Probabilistic methods are usually used to discriminate between the clutter and the target measurements. Clutter measurement density is an important parameter in this process. The values of the clutter measurement density in the surveillance space are rarely known a priori, and are usually estimated using sensor data and track information. A novel approach is presented and evaluated for estimating the values of clutter measurement density, which significantly enhances target tracking. Simulation results validate this approach.
International Conference on Information Fusion, Jul 5, 2016
The target tracking using multistatic passive radar in a digital audio/video broadcast (DAB/DVB) ... more The target tracking using multistatic passive radar in a digital audio/video broadcast (DAB/DVB) network with uncertain illuminators of opportunity faces two main challenges: one is the three-dimensional (3-D) data association problem among the target-measurement-illuminator due to the uncertainty of illuminators of opportunity; the other is that only the bistatic range and range-rate measurements are available since the angle information is unavailable or of very poor quality. In this paper, the authors propose two tracking algorithms directly in a two dimensional (2-D) Cartesian coordinates with the capability of false track discrimination (FTD) using the probability of target existence: the modified joint integrated data association (MJIPDA) and the sequential processing joint integrated data association (SP-JIPDA). The MJIPDA algorithm is enhanced based on the existing MJPDA algorithm by using the probability of target existence for FTD. The SP-JIPDA algorithm sequentially operates the JIPDA tracker to update each track for each illuminator with all the measurements in the common measurement set at each time. A simulation study is performed to verify the validity of both algorithms in this multistatic passive radar system.
International Conference on Information Fusion, Jul 9, 2012
This paper presents an effective recursive tracking method for the single mobile emitter using co... more This paper presents an effective recursive tracking method for the single mobile emitter using correlated TDOA measurements. In multiple-sensor environment the time difference of arrival (TDOA) measurements are correlated. We consider only the case of single emitter tracking without data association issues, e.g., no missed detection or false measurements. We describe a scenario of three static receivers with synchronous TDOA measurements. In this situation, the TDOA measurements are correlated. We propose a decorrelated TDOA measurement structure for optimal estimation. Each TDOA measurement defines a hyperbola-like region, representing possible emitter locations. This likelihood function is approximated via Gaussian mixture, making the algorithm a dynamic bank of variable number of Kalman filters. The performance is evaluated using Monte Carlo simulations, and compared with the Cramer-Rao Lower Bound (CRLB). I.
International Conference on Information Fusion, Jul 9, 2013
ABSTRACT Data association attempts to discriminate between the target and the clutter measurement... more ABSTRACT Data association attempts to discriminate between the target and the clutter measurements, usually calculating the posterior probabilities of measurement origins. The clutter (spurious) measurements are random and (we presume) follow the Poisson distribution. The Poisson distribution is non-homogeneous and is parametrized by intensity (the clutter measurement density). The clutter measurement density is almost always a priori unknown, and is often non stationary. Here we propose a measurement oriented clutter density estimator with probability hypothesis density (PHD) filtering which integrates information from single-scan spatial clutter density estimator, and can follow and smooth non-stationary clutter information.
We propose an effective method of extracting targets from a region-of-interest (ROI) in infrared ... more We propose an effective method of extracting targets from a region-of-interest (ROI) in infrared images by using a 2-D histogram, considering intensity values and distance values from a center of the ROI. Existing approaches for extracting targets have utilized only intensity values of pixels or an analysis of a 1-D histogram of intensity values. Because the 1-D histogram has mixed bins containing false-negative bins from the target region as well as false-positive bins from the background region, it is difficult to extract target regions effectively due to the mixed bins. In order to solve the problem of the mixed bins, we propose a novel 2-D histogrambased approach for extracting targets. Experimental results have shown that the proposed method achieves better performance of extracting targets than existing methods under various environments, such as target regions with irregular intensities, dim targets, and cluttered backgrounds.
Multi-target tracking in clutter, assuming linear target trajectory propagation and linear target... more Multi-target tracking in clutter, assuming linear target trajectory propagation and linear target measurement equation, naturally leads to a Gaussian mixture (GM) target tracking solution. This study examines and compares two prominent methods that use the GMs: the probability hypothesis density and the integrated track splitting. Both are recursive Bayes methods and both incorporate the false track discrimination capabilities. They are represented in the form of GM target density filters. The modelling assumptions are translated in the algorithmic requirements. The authors compare the algorithms on the basis of these requirements with the future work indicated to reconcile algorithms and requirements.
In situations with a significant number of targets in mutual proximity (close to each other), opt... more In situations with a significant number of targets in mutual proximity (close to each other), optimal multitarget data association approach suffers from the numerical explosion. This severely limits the applicability, i.e., the number of close targets that may be reliably tracked. We propose an iterative implementation of the joint integrated probabilistic data association (JIPDA) which allows a performance/computation resources tradeoff. This approach can also be incorporated into joint integrated track splitting (JITS). The iterations start with the single target integrated probabilistic data association (IPDA) and each subsequent iteration improves the approximation towards JIPDA, reaching the optimal multitarget solution within a finite number of iterations.
The point target assumption which allows each target to generate at most one measurement at each ... more The point target assumption which allows each target to generate at most one measurement at each scan is widely used in target tracking field. However, in come tracking scenarios, one target can generate many measurements due to high sensor resolution which give rise to the multiple detection problem. Traditional algorithms get poor performances since each data association event considers only one measurement as target detection to estimate target state. The measurement partition method which forms all possible combinations of target originated measurements is designed for multiple detection problem. In this paper, joint integrated track splitting (JITS) tracker and measurement partition method are combined to generate a new structure, called multiple detection joint integrated track splitting (MD-JITS) tracker, to extract and utilize the target motion information contained in the measurements more effectively. Compared with some existing filters, this filter achieves better tracking performance for automatic false track discrimination (FTD).
We investigate two solutions for single target tracking in clutter using a high pulse repetition ... more We investigate two solutions for single target tracking in clutter using a high pulse repetition frequency (HPRF) radar. With the HPRF radar, each ambiguous range measurement corresponds to a set of unambiguous range measurements. Both algorithms enable false track discrimination (FTD) using the probability of target existence. One approach adapts the Gaussian mixture measurement likelihood-integrated track splitting to the problem, where each measurement likelihood is approximated by a Gaussian mixture. The other approach is an integration of previously published multiple models approach with the probability of target existence paradigm. Here each model is a separate probabilistic data association filter. In both cases the track trajectory probability density function is a Gaussian mixture.
A radar system for tracking ground vehicle targets to realize adaptive cruise control requires an... more A radar system for tracking ground vehicle targets to realize adaptive cruise control requires an accurate vehicle tracking filter. Especially, for ground vehicle tracking, one has to consider the case in which the radar measurements are affected by glint noises generated by the targets located near the observer vehicle. Furthermore, this vehicle tracking should be performed in cluttered environments. This paper presents a combined algorithm that consists of integrated probabilistic data association (IPDA), and an interacting multiple model (IMM) algorithm for ground target tracking in clutter and target glint. Performance of the proposed algorithm is tested and verified by a series of computer simulation runs.
This study presents a complete algorithm for single target tracking in clutter, which addresses s... more This study presents a complete algorithm for single target tracking in clutter, which addresses simultaneously: nonlinear 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 nonlinear 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.
ABSTRACT We consider passive surveillance using time and frequency difference of arrival signals ... more ABSTRACT We consider passive surveillance using time and frequency difference of arrival signals received by mobile receiver pairs. Signals received by a pair of receivers are correlated in time and frequency, followed by a detection process. In addition to the target (emitter) measurements, we may also create a number of spurious detections in each scan. This paper considers local (distributed) tracking using these measurements, with the main purpose of eliminating spurious measurements and enhancing the emitter detection.
This paper presents an algorithm for multistatic target tracking in clutter, using only range dif... more This paper presents an algorithm for multistatic target tracking in clutter, using only range difference information (neither bearing nor Doppler information are assumed available). Presence of false tracks, Data association issues as well as the nonlinear measurement equation makes this a challenging problem. This paper proposes a solution to this problem by using the Gaussian Mixture Measurement likelihood — Integrated Track Splitting algorithm.
The authors extend the smoothing integrated probabilistic data association algorithm to multi-tar... more 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.
Tracking an unknown target in noisy environment is difficult especially when the target is maneuv... more Tracking an unknown target in noisy environment is difficult especially when the target is maneuvering and has unknown trajectory. Smoother uses measurements from future scans to estimate the target state in past scan. This requires the fusion of forward and backward prediction. However, due to uncertain target motions and low detection probabilities, backward prediction could not associate with forward prediction which results in inequitable fusion pair and thus, smoothing performance could not be improved. To cope with these difficulties, the proposed algorithm modifies the fixed-lag smoothing data association based on the integrated probabilistic data association (IPDA) algorithm and a new algorithm called modified smoothing IPDA (MSIPDA) is developed. MSIPDA utilizes two IPDA filters to obtain forward IPDA (fIPDA) track and backward (bIPDA) track estimation in each scan. Each fIPDA prediction generates multiple fusion pairs in association with bIPDA multi-track prediction. These fusion pairs are created in the form of components. As a result, multiple smoothing components are formed with their smoothing component data association probabilities for computing MSIPDA track components state estimate. In addition, the smoothing data association probabilities upgrade the forward track which makes the forward track more powerful for target tracking in clutter. The numerical assessment of MSIPDA is verified using simulations. The result shows significant false track discrimination performance in comparison to existing algorithms.
International Journal of Control Automation and Systems, Feb 1, 2008
An interacting multiple model (IMM) estimation algorithm based on the mixing of the predicted sta... more An interacting multiple model (IMM) estimation algorithm based on the mixing of the predicted state estimates is proposed in this paper for a right continuous jump-linear system model different from the left-continuous system model used to develop the existing IMM algorithm. The difference lies in the modeling of the mode switching time. Performance of the proposed algorithm is compared numerically with that of the existing IMM algorithm for noisy system identification. Based on the numerical analysis, the proposed algorithm is applied to target tracking with a large sampling period for performance comparison with the existing IMM.
A Gaussian Mixture (GM) target tracking solution is a natural consequence of the multi-target tra... more A Gaussian Mixture (GM) target tracking solution is a natural consequence of the multi-target tracking in clutter, given a linear target trajectory propagation and a linear target measurement equation. We examine and compare two prominent GM target trackers: the Multi Hypothesis Tracking (MHT) and the Integrated Track Splitting. Both incorporate the false track discrimination capabilities, enabling automatic target tracking in the presence of clutter measurements and missed detections.
Multistatic tracking involves using non-collocated transmitters and receivers to track the target... more Multistatic tracking involves using non-collocated transmitters and receivers to track the targets. In this paper we also assume no angle information. This paper proposes an approach using Gaussian Mixture representation of measurement uncertainty. An arbitrary number of receivers can be used using this approach. Here we limit ourselves to the track estimation issues, i.e. we assume no clutter measurement, and imperfect target detection at the receivers. Simulation results are presented which vindicate this approach.
IEEE Transactions on Aerospace and Electronic Systems, Apr 1, 2011
In a surveillance situation the origin of each measurement is uncertain. Each measurement may be ... more In a surveillance situation the origin of each measurement is uncertain. Each measurement may be a false (clutter) measurement, or it may be a target detection. Probabilistic methods are usually used to discriminate between the clutter and the target measurements. Clutter measurement density is an important parameter in this process. The values of the clutter measurement density in the surveillance space are rarely known a priori, and are usually estimated using sensor data and track information. A novel approach is presented and evaluated for estimating the values of clutter measurement density, which significantly enhances target tracking. Simulation results validate this approach.
International Conference on Information Fusion, Jul 5, 2016
The target tracking using multistatic passive radar in a digital audio/video broadcast (DAB/DVB) ... more The target tracking using multistatic passive radar in a digital audio/video broadcast (DAB/DVB) network with uncertain illuminators of opportunity faces two main challenges: one is the three-dimensional (3-D) data association problem among the target-measurement-illuminator due to the uncertainty of illuminators of opportunity; the other is that only the bistatic range and range-rate measurements are available since the angle information is unavailable or of very poor quality. In this paper, the authors propose two tracking algorithms directly in a two dimensional (2-D) Cartesian coordinates with the capability of false track discrimination (FTD) using the probability of target existence: the modified joint integrated data association (MJIPDA) and the sequential processing joint integrated data association (SP-JIPDA). The MJIPDA algorithm is enhanced based on the existing MJPDA algorithm by using the probability of target existence for FTD. The SP-JIPDA algorithm sequentially operates the JIPDA tracker to update each track for each illuminator with all the measurements in the common measurement set at each time. A simulation study is performed to verify the validity of both algorithms in this multistatic passive radar system.
International Conference on Information Fusion, Jul 9, 2012
This paper presents an effective recursive tracking method for the single mobile emitter using co... more This paper presents an effective recursive tracking method for the single mobile emitter using correlated TDOA measurements. In multiple-sensor environment the time difference of arrival (TDOA) measurements are correlated. We consider only the case of single emitter tracking without data association issues, e.g., no missed detection or false measurements. We describe a scenario of three static receivers with synchronous TDOA measurements. In this situation, the TDOA measurements are correlated. We propose a decorrelated TDOA measurement structure for optimal estimation. Each TDOA measurement defines a hyperbola-like region, representing possible emitter locations. This likelihood function is approximated via Gaussian mixture, making the algorithm a dynamic bank of variable number of Kalman filters. The performance is evaluated using Monte Carlo simulations, and compared with the Cramer-Rao Lower Bound (CRLB). I.
International Conference on Information Fusion, Jul 9, 2013
ABSTRACT Data association attempts to discriminate between the target and the clutter measurement... more ABSTRACT Data association attempts to discriminate between the target and the clutter measurements, usually calculating the posterior probabilities of measurement origins. The clutter (spurious) measurements are random and (we presume) follow the Poisson distribution. The Poisson distribution is non-homogeneous and is parametrized by intensity (the clutter measurement density). The clutter measurement density is almost always a priori unknown, and is often non stationary. Here we propose a measurement oriented clutter density estimator with probability hypothesis density (PHD) filtering which integrates information from single-scan spatial clutter density estimator, and can follow and smooth non-stationary clutter information.
We propose an effective method of extracting targets from a region-of-interest (ROI) in infrared ... more We propose an effective method of extracting targets from a region-of-interest (ROI) in infrared images by using a 2-D histogram, considering intensity values and distance values from a center of the ROI. Existing approaches for extracting targets have utilized only intensity values of pixels or an analysis of a 1-D histogram of intensity values. Because the 1-D histogram has mixed bins containing false-negative bins from the target region as well as false-positive bins from the background region, it is difficult to extract target regions effectively due to the mixed bins. In order to solve the problem of the mixed bins, we propose a novel 2-D histogrambased approach for extracting targets. Experimental results have shown that the proposed method achieves better performance of extracting targets than existing methods under various environments, such as target regions with irregular intensities, dim targets, and cluttered backgrounds.
Multi-target tracking in clutter, assuming linear target trajectory propagation and linear target... more Multi-target tracking in clutter, assuming linear target trajectory propagation and linear target measurement equation, naturally leads to a Gaussian mixture (GM) target tracking solution. This study examines and compares two prominent methods that use the GMs: the probability hypothesis density and the integrated track splitting. Both are recursive Bayes methods and both incorporate the false track discrimination capabilities. They are represented in the form of GM target density filters. The modelling assumptions are translated in the algorithmic requirements. The authors compare the algorithms on the basis of these requirements with the future work indicated to reconcile algorithms and requirements.
In situations with a significant number of targets in mutual proximity (close to each other), opt... more In situations with a significant number of targets in mutual proximity (close to each other), optimal multitarget data association approach suffers from the numerical explosion. This severely limits the applicability, i.e., the number of close targets that may be reliably tracked. We propose an iterative implementation of the joint integrated probabilistic data association (JIPDA) which allows a performance/computation resources tradeoff. This approach can also be incorporated into joint integrated track splitting (JITS). The iterations start with the single target integrated probabilistic data association (IPDA) and each subsequent iteration improves the approximation towards JIPDA, reaching the optimal multitarget solution within a finite number of iterations.
The point target assumption which allows each target to generate at most one measurement at each ... more The point target assumption which allows each target to generate at most one measurement at each scan is widely used in target tracking field. However, in come tracking scenarios, one target can generate many measurements due to high sensor resolution which give rise to the multiple detection problem. Traditional algorithms get poor performances since each data association event considers only one measurement as target detection to estimate target state. The measurement partition method which forms all possible combinations of target originated measurements is designed for multiple detection problem. In this paper, joint integrated track splitting (JITS) tracker and measurement partition method are combined to generate a new structure, called multiple detection joint integrated track splitting (MD-JITS) tracker, to extract and utilize the target motion information contained in the measurements more effectively. Compared with some existing filters, this filter achieves better tracking performance for automatic false track discrimination (FTD).
We investigate two solutions for single target tracking in clutter using a high pulse repetition ... more We investigate two solutions for single target tracking in clutter using a high pulse repetition frequency (HPRF) radar. With the HPRF radar, each ambiguous range measurement corresponds to a set of unambiguous range measurements. Both algorithms enable false track discrimination (FTD) using the probability of target existence. One approach adapts the Gaussian mixture measurement likelihood-integrated track splitting to the problem, where each measurement likelihood is approximated by a Gaussian mixture. The other approach is an integration of previously published multiple models approach with the probability of target existence paradigm. Here each model is a separate probabilistic data association filter. In both cases the track trajectory probability density function is a Gaussian mixture.
A radar system for tracking ground vehicle targets to realize adaptive cruise control requires an... more A radar system for tracking ground vehicle targets to realize adaptive cruise control requires an accurate vehicle tracking filter. Especially, for ground vehicle tracking, one has to consider the case in which the radar measurements are affected by glint noises generated by the targets located near the observer vehicle. Furthermore, this vehicle tracking should be performed in cluttered environments. This paper presents a combined algorithm that consists of integrated probabilistic data association (IPDA), and an interacting multiple model (IMM) algorithm for ground target tracking in clutter and target glint. Performance of the proposed algorithm is tested and verified by a series of computer simulation runs.
This study presents a complete algorithm for single target tracking in clutter, which addresses s... more This study presents a complete algorithm for single target tracking in clutter, which addresses simultaneously: nonlinear 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 nonlinear 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.
ABSTRACT We consider passive surveillance using time and frequency difference of arrival signals ... more ABSTRACT We consider passive surveillance using time and frequency difference of arrival signals received by mobile receiver pairs. Signals received by a pair of receivers are correlated in time and frequency, followed by a detection process. In addition to the target (emitter) measurements, we may also create a number of spurious detections in each scan. This paper considers local (distributed) tracking using these measurements, with the main purpose of eliminating spurious measurements and enhancing the emitter detection.
This paper presents an algorithm for multistatic target tracking in clutter, using only range dif... more This paper presents an algorithm for multistatic target tracking in clutter, using only range difference information (neither bearing nor Doppler information are assumed available). Presence of false tracks, Data association issues as well as the nonlinear measurement equation makes this a challenging problem. This paper proposes a solution to this problem by using the Gaussian Mixture Measurement likelihood — Integrated Track Splitting algorithm.
The authors extend the smoothing integrated probabilistic data association algorithm to multi-tar... more 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.
Tracking an unknown target in noisy environment is difficult especially when the target is maneuv... more Tracking an unknown target in noisy environment is difficult especially when the target is maneuvering and has unknown trajectory. Smoother uses measurements from future scans to estimate the target state in past scan. This requires the fusion of forward and backward prediction. However, due to uncertain target motions and low detection probabilities, backward prediction could not associate with forward prediction which results in inequitable fusion pair and thus, smoothing performance could not be improved. To cope with these difficulties, the proposed algorithm modifies the fixed-lag smoothing data association based on the integrated probabilistic data association (IPDA) algorithm and a new algorithm called modified smoothing IPDA (MSIPDA) is developed. MSIPDA utilizes two IPDA filters to obtain forward IPDA (fIPDA) track and backward (bIPDA) track estimation in each scan. Each fIPDA prediction generates multiple fusion pairs in association with bIPDA multi-track prediction. These fusion pairs are created in the form of components. As a result, multiple smoothing components are formed with their smoothing component data association probabilities for computing MSIPDA track components state estimate. In addition, the smoothing data association probabilities upgrade the forward track which makes the forward track more powerful for target tracking in clutter. The numerical assessment of MSIPDA is verified using simulations. The result shows significant false track discrimination performance in comparison to existing algorithms.
Uploads
Papers by Taek Song