Tracking of multiple maneuvering targets using multiscan JPDA and IMM filtering

The Set JPDA (SJPDA) filter is a recently developed multi-target tracking filter that utilizes the relationbetween the density of a random finite set and the ordinary density of a state vector to improve on the JointProbabilistic Data Association (JPDA) filter. One advantage to the filter is the improved accuracy of theGaussian approximations of the JPDA, which results in avoidance of track coalescence. Another advantage is animproved estimation accuracy in terms of a measure which disregards target identity. In this paper we extend thefilter to also consider multiple motion models. As a basis for the extension we use the Interacting Multiple Model(IMM) algorithm. We derive three alternative filters that we jointly refer to as Set IMMJPDA (SIMMJPDA).They are based on two alternative descriptions of the IMMJPDA filter. In the paper, we also present simulationresults for a two-target tracking scenario, which show improved tracking performance for the Set IMMJPDAfilter when evaluated with a measure that disregards target identity.

Variable Structure Multiple Model (VSMM) estimation generalizes Multiple Model (MM) estimation by assuming that the set of models used for MM estimation is time varying. By using VSMM estimation, a large model set which may cover all possible target maneuvers can be used without significant increase of computational load, while maintaining a reasonable estimation accuracy. Various implementable VSMM algorithms, like Model Group Switching (MGS), Likely Model Set (LMS) and Minimal Sub-Model-Set Switching (MSMSS) using the Interacting Multiple Model (IMM) algorithm with a sub-model-set adaptation logic have appeared in the recent literature. However, the use of these algorithms for tracking maneuvering target in clutter has not been explored. In presence of clutter, one need to use data association technique to differentiate target originated measurement from clutter. The probabilistic data association (PDA) has been popularly adopted to many algorithms for tracking in clutter. In this paper, we integrate PDA technique with MSMSS and propose a VSIMM-PDA algorithm for maneuvering target tracking in clutter. A new grating technique to account for potential model errors is used. A numeric example via multiple Monte Carlo runs, which compares the performance of the new algorithm to a standard IMM-PDA in terms of Root Mean Squared error (RMS) and percentage of track loss, is presented.© (2003) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

An adaptive channel estimation scheme, exploiting the over-sampled complex exponential basis expansion model (CE-BEM), is presented for doubly-selective channels where we track the BEM coefficients via a multiple model approach. In the past work the number of BEM coefficients used to model the doubly-selective channels for channel estimation has been based on an upperbound on the channel Doppler spread. Higher the Doppler spread, more the number of BEM coefficients leading to higher channel estimation variance. In this paper we propose to use a multiple model framework where several candidate Doppler spread values are used to cover the range from zero to an upperbound, leading to multiple CE-BEM channel models, each corresponding to an assumed value of the Doppler spread. Subsequently the well- known interacting multiple model (IMM) algorithm is used for symbol detection based on multiple state-space models corresponding to the multiple estimated channels. A simulation example is presented to illustrate the proposed approach.

Multiple target tracking (MTT) has received much attention recently for various applications in the military as well as the Strategic Defense Initiative areas. Data association is one of the critical computation in MTT problems, because erroneous data associations often result in lost tracks. The joint probabilistic data association (JPDA) algorithm is a good approach to solve the data association problem. However, the computation complexity of this algorithm increases rapidly with the number of targets and radar returns. Neural networks have been considered to approximate the JPDA and ease the computation burden through the parallel processing. In this paper, we propose a neural network data association (NNDA) algorithm for the solution of the data association problems. Simulation results show the following three controversial issues: first, NNDA can track multiple targets with performance compatible with JPDA. Second, when the prediction filter can not provide an enough accurate prediction data ( This sometimes occurs when the prediction model can not match the tracking environment precisely enough), the neural computation provides better performace than JPDA. Third, the performance of NNDA is not affected by the number while that of JPDA degrades with the increase of target number. As a whole, this paper presents a neural network which not only possesses the intrinsic ability of parallel computation but also provides conditionally better tracking performance than JPDA.

A single-sensor, single-target mixture reduction (MR) data association algorithm is extended for use in multisensor, multitarget tracking situations. MR is extended for tracking an arbitrary number of targets using an arbitrary number of sensors under the assumption that the sensor measurement errors are independent across sensors. Like the single-sensor, single-target MR algorithm, which gives better performance than the probabilistic data association (PDA) filter, the multisensor, multitarget MR extensions give similar improvements in performance compared to the joint PDA (JPDA) and multisensor JPDA (MSJPDA) algorithms. Further, in the formulations for the multisensor and multitarget MR algorithms, the equations for the calculation of the data association probabilities have been put in the same form as for the JPDA, thus allowing previously developed fast JPDA computational techniques to be applicable. N the problem of tracking targets in random clutter, the optimal Bayesian solution leads to Gaussian mixture distributions that consist of an exponentially increasing number of components.1 In practice, the number of components in each mixture is kept in check by approximating it with fewer components. The nearest neighbor (NN) approximation reduces each target's mixture to its largest component.13 The joint probabilistic data association (JPDA) approximation merges the components of each target's mixture into a single Gaussian component.1'4 However, when there are several significant, well-spaced components in the original mixture, important information may be lost by using the NN or JPDA methods. A more flexible approximation method for single-sensor, single-target tracking, known as mixture reduction (MR), has been proposed by Salmond.5 This MR algorithm preserves the mean and covariance of the original mixture of components, successively merging components until the number of components is reduced to some userspecified limit. In this paper, we first review Salmond's MR algorithm for single-sensor tracking of a single target in clutter. We then derive extensions to Salmond's MR algorithm for use in tracking multiple targets and when there are measurements available from multiple sensors. Just as the single-sensor, single-target MR algorithm gives better performance5 than the probabilistic data association (PDA) filter, the multisensor, multitarget MR extensions are shown to give similar improvements in performance as compared to the JPDA and multisensor JPDA (MSJPDA)6 algorithms. Extension of the MR algorithm for tracking of multiple targets is necessary for use in realistic tracking scenarios where there are usually many targets that need to be tracked. Moreover, many current tracking systems have been designed with multiple sensors, thus requiring the further extension to a multisensor algorithm. The paper is organized as follows. In Sec. II, the target dynamics and measurement equations are defined. We review the singlesensor, single-target MR algorithm5 in Sec. Ill, and we extend the algorithm to the multitarget and multisensor cases in Sec. IV and V, respectively. In Sec. VI, simulation results evaluating the performances of the MR extensions are presented. Finally, concluding remarks are given in Sec. VII.

Distributed asynchronous multiple sensor fusion with nonlinear multiple models

In this paper, an algorithm is proposed for tracking a highly maneuvering target in the presence of clutter by fusion of multiple detection radar and infrared sensor. It is presented by combining interactive multiple model (IMM) and multiple detection probabilistic data association filter (MDPDAF) in a multi-sensor (radar and infrared) scenario known as IMM/MS-MDPDAF algorithm. Multiple radar detection could be due to multipath or extended nature of the target such that in each scan there would be more than one valid target detection. In the presence of multiple radar detection there is uncertainty in data association in addition to the measurements origin (target or false alarm), in the number of measurements arising from the target. Accordingly MDPDAF algorithm that considers both uncertainties is developed instead of the standard PDAF algorithm which only considers the uncertainty in the measurements origin. The simulation has shown that in the presence of multiple radar detection, tracking with the proposed algorithm that considers both uncertainties has considerable improvement compared to the IMM/MSPDAF algorithm that only considers uncertainty in the measurements origin.

Estimation of 3D piecewise coordinated turn target maneuvers with unknown turning rates is considered. The State Dependent Differential-Difference Riccati Equation (SDDRE) and the Interacting Multiple Model (IMM) approaches are compared. The comparison is kept as simple as possible and effort is made to perform it on a common basis. The SDDRE uses a non-linear modeling of the target's equations of motion that takes into account the rotation of the velocity and acceleration vectors as the target maneuvers. The IMM implements two models: a constant acceleration target model and a simple uncoupled coordinated turn maneuver model using the average-expected turn rate. The simulated trajectory is a combination of constant velocity trajectory segments and coordinated turn segments. The simulations show advantages and disadvantages of each of the approaches. It is demonstrated that the SDDRE approach filter is not sensitive to the trajectory orientation and that it has faster transient thus demonstrates a kind of adaptivity to the turning rate level. The IMM approach filter has the advantages that when the target trajectory is close to one of the hypotheses it has better performance than the SDDRE based filter.

Mechanically steered scanning radars receive measurements continuously while sweeping the surveillance region. However, most target tracking algorithms, like the multiple hypothesis tracker (MHT) and the joint probabilistic data association (JPDA) techniques, wait for the end of a scan in order to process the measurements and to estimate targets states. This is due to the fundamental assumption of one-to-one association between tracks and measurements and the 360° physical limit of a scan. Associating measurements to initialized tracks and filtering at the end of a complete scan may cause significant delays in target state update. In addition, association may become imperfect due to longer intervals between updates. This issue becomes significant when tracking high-speed targets with low scan rate sensors as in the airborne early warning (AEW) system. In this paper, we present a new dynamic sector processing algorithm using two-dimensional (2-D) assignment for scanning radars that report measurements within the duration of a scan. The full scan is dynamically and adaptively divided into sectors, which could be as small as a single detection, depending on the scanning rate, sparsity of targets, and required target state update speed. Measurement-to-track association, filtering, and target state update are done dynamically while sweeping from one region to another, i.e., continuous track update, limited only by the intermeasurement interval, becomes possible. The proposed algorithm offers low latency while maintaining estimation accuracy in track updates as well as efficient utilization of computational resources compared with standard frame-based tracking algorithms. Experimental results based on rotating radars demonstrate the advantages of the proposed technique.

Interacting multiple model (IMM) filtering and multiple hypothesis tracking (MHT) represent the most accurate methods currently available for tracking multiple maneuvering targets in cluttered environments. Although these methods are complex, modern computational capabilities make their combined implementation feasible for modern tracking systems. This paper discusses alternative approaches for developing a combined IMM/MHT tracking system and describes specific implementations that have been developed for radar and IRST applications. Simulation results for both radar and IRST systems are presented. Track maintenance and accuracy performance for the IMM/MHT system is compared with that obtained from an MHT tracker using a conventional filtering approach. Results indicate that the improvements in data association derived from the use of IMM filtering with MHT may be comparable to the well known IMM improvements in track accuracy. The addition of multiple filter models to MHT data association has the potential to significantly increase computational requirements. Thus, several compromises, described in this paper, have been developed in order to assure computational feasibility. Preliminary estimates of computational requirements are given in order to demonstrate implementation feasibility.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

For multi-target tracking, the IMM (interactive multiple model) algorithm has been proposed. The IMM is expected to reduce tracking errors for both nonmaneuvering and maneuvering targets. However, the IMM requires a heavy computational burden, because it utilizes multiple Kalman filters in parallel. On the other hand, the Kalman filter with a turning acceleration estimator, which can adapt a maneuvering target, has been proposed. The Kalman filter with a turning acceleration estimator is a single two-stage type filter and has a problem of setting threshold for the maneuver detector. In this paper, we propose the hybrid filter with a constant-velocity (CV) filter and a turning acceleration estimation filter. The proposed filter does not require a maneuver detector, because it integrates the outputs of two filters using the likelihood of each filter. And its computational requirement is smaller than the IMM, since it consists of only two Kalman based filters. The proposed method can prevent deteriorating tracking accuracy by reducing the risk of maneuver misdetection when a target maneuvers. We evaluate the performance of the proposed filter by computer simulation, and show the effectiveness of the proposed filter, comparing with the conventional Kalman filter and the two-stage Kalman filter.

A new method, M/sup 3/JPDA (multiple maneuver model JPDA), for maneuvering target tracking in a dense environment is the extension of the conventional JPDA (joint probabilistic data association). In the M/sup 3/JPDA, the maneuver acceleration is assumed to be limited to a time invariant set of discrete values. But, in the M/sup 3/JPDA, there is a case that track maintenance capability of two crossing targets is inferior to that of the JPDA. Therefore, some adaptive techniques are proposed to improve the performance of the M/sup 3/JPDA. The performance of this method is evaluated in terms of tracking success rates by computer simulation.

When conducting performance assessment of multi-target tracking algorithms with a realistic computer simulation or real-world sensor data, good track-to-truth assignment (TTA) is a critical component of any meaningful assessment. The presence of kinematic data for small objects that may not be observed by the sensor in the truth data poses serious challenges to the TTA. The assignment process is further complicated by the presence of sensor biases. In this paper, TTA in the presence of small objects and sensor biases is considered. When the target density is high and some objects are small, correctly assigning tracks to truth objects is challenging, and with the addition of sensor biases, the problem of correctly assigning tracks to truth objects is impossible without mitigation. The high rate of false switches of tracks between true objects greatly hinders the performance assessment of the target tracking system. In this paper, these challenges are addressed by adding probability of tracking to a probabilistic data association (PDA) technique for TTA. The computational algorithms for the implementation of the PDA technique are presented along with simulation results that confirm the effectiveness of the PDA approach in accurately estimating the sensor biases, and reducing the artificial track switches and ambiguity in the TTA.

The conventional interacting multiple models (IMM) approach for a hybrid system under the Gaussian assumption is limited for most real applications due to the noisy measurements often being in the presence of outliers. This paper aims at accommodating the IMM approach to the non‐Gaussian cases where outliers exist. In the proposed IMM algorithm, the Student‐t distribution is used to model the non‐Gaussian measurement noise. At the interaction step, the mixed statistics of the noise parameter under a Bayesian framework are obtained via a Gamma approximation and a recently reported moments matching method. To address the state noise‐coupled intractability in Bayesian filtering, a variational Bayesian method is utilized to approximate the posterior distributions of the noise and state recursively. The proposed algorithm is tested with a maneuvering target tracking example and is shown to be robust to the outliers.

Particle filters can provide better results than Kalman filter based algorithms in the problems where the nonlinear elements are dominant. In this study the problem of tracking an airborne platform using radar measurements, which has nonlinearities both in measurement and system models is considered. Different models are used for different target maneuvers; extended Kalman filter (EKF), interacting multiple model (IMM) structure and multiple model particle filter are compared on a sample scenario.