Multi-sensor Joint Probabilistic Data Association Research Articles

Distributed Unmanned Aerial Vehicle Positioning for Multitarget Tracking with Bearing-Only Information

This paper proposes a distributed positioning algorithm for a swarm of unmanned aerial vehicles (UAVs) to track multiple moving targets with bearing-only measurements. An approximate performance metric is first derived that can be used in position determination based on the properties of multisensor joint probabilistic data association (JPDA) filter. A fully distributed position planning algorithm using incremental optimization strategy is then proposed for tracking multiple moving targets. Simulation examples with comparison results validate the effectiveness of the proposed algorithm.

Sequential Multi-Sensor JPDA for Target Tracking in Passive Multi-Static Radar With Range and Doppler Measurements

Target tracking in passive multi-static radar (PMSR) with bistatic range and Doppler frequency measurements from multiple transmit–receive pairs is gaining increasing interest. For the data association problem in this scenario, the parallel architecture of a multi-sensor joint probabilistic data association (P-MSJPDA) filter has been significantly investigated. As an alternative architecture, the sequential MSJPDA (S-MSJPDA) is rarely discussed in PMSR. In this paper, we evaluate the behaviors of S-MSJPDA in PMSR target tracking with bistatic range and Doppler frequency measurements. A comprehensive comparison between the S-MSJPDA and the P-MSJPDA in PMSR is provided. It can be found from the analysis that S-MSJPDA outperforms its parallel counterpart in terms of computational efficiency, given an acceptable degradation in position accuracy. The S-MSJPDA is further applied to an experimental passive multi-static radar for aircrafts tracking. The real data results obtained are rather close to the true trajectories of the targets. This demonstrates that the S-MSJPDA has great potentials in PMSR target tracking.

Joint spatial registration and multi-target tracking using an extended PM-CPHD filter

An extended product multi-sensor cardinalized probability hypothesis density (PM-CPHD) filter for spatial registration and multi-target tracking (MTT) is proposed. The number and states of targets and the biases of sensors are jointly estimated by this method without the data association. Monte Carlo (MC) simulation results show that the proposed method (i) outperforms, although computationally more expensive than, the extended multi-sensor PHD filter which has been proposed for joint spatial registration and MTT; (ii) outperforms the multi-sensor joint probabilistic data association (MSJPDA) filter which is also extended in this study for joint spatial registration and MTT when the clutter is relatively dense.

Joint spatial registration and multi-target tracking using an extended probability hypothesis density filter

An extended probability hypothesis density filter for translational measurement registration and multi-target tracking (MTT) is proposed. The number and states of the targets and the biases of the sensors are jointly estimated by this method without the data association. The sequential Monte Carlo method is used to implement the proposed algorithm considering non-linear and non-Gaussian conditions. Monte Carlo simulation results show that the proposed method (i) outperforms, although computationally a little more expensive than, the standard PHD filter which does not involve the process of spatial registration; (ii) outperforms the multi-sensor joint probabilistic data association (MSJPDA) filter which is also extended in this study for joint spatial registration and MTT when the clutter is relatively dense.

Approximate stochastic dynamic programming for sensor scheduling to track multiple targets

The problem of sensor scheduling is to select the number and combination of sensors to activate over time. The goal is usually to trade off tracking performance and sensor usage. We formulate a version of this problem involving multiple targets as a partially observable Markov decision process, and use this formulation to develop a nonmyopic sensor-scheduling scheme. Our scheme integrates sequential multisensor joint probabilistic data association and particle filtering for belief-state estimation, and use a simulation-based Q-value approximation method called completely observable rollout for decision making. We illustrate the effectiveness of our approach by an example with multiple sensors activated simultaneously to track multiple targets. We also explore the trade-off between tracking error and sensor cost using our nonmyopic scheme.

Distributed interacted multisensor joint probabilistic data association algorithm based on D-S theory

In order to resolve the multisensor multiplied maneuvering target tracking problem, this paper presents a distributed interacted multiple model multisensor joint probabilistic data association algorithm (DIMM-MSJPDA). First of all, the interacted multiple model joint probabilistic data association algorithm is applied to each sensor, and then the state estimation, estimation covariance, model probability, combined innovation, innovation covariance are delivered to the fusion center. Then, the tracks from each sensor are correlated and the D-S evidence theory is used to gain the model probability of an identical target. Finally, the ultimate state estimation of each target is calculated according to the new model probability, and the state estimation is transmitted to each sensor. Simulations are designed to test the tracking performance of DIMM-MSJPDA algorithm. The results show that the use of DIMM-MSJPDA algorithm enables the distributed multisensor system to track multiplied maneuvering targets and its tracking performance is much better than that of IMMJPDA algorithm.

Variance estimation and ranking of target tracking position errors modeled using Gaussian mixture distributions

In this paper, variance estimation and ranking methods are developed for stochastic processes modeled by Gaussian mixture distributions. It is shown that the variance estimate from a Gaussian mixture distribution has the same properties as the variance estimate from a single Gaussian distribution based on a reduced number of samples. Hence, well-known tools for variance estimation and ranking of single Gaussian distributions can be applied to Gaussian mixture distributions. As an application example, we present optimization of sensor processing order in the sequential multi-target multi-sensor joint probabilistic data association (MSJPDA) algorithm.

A Comparison of Parallel and Sequential Implementations of a Multisensor Multitarget Tracking Algorithm

Parallel and sequential implementations of the Multisensor Joint Probabilistic Data Association (MSJPDA) tracking algorithm are analyzed and compared. Two non-simulation techniques for comparison of multisensor probabilistic data association filters are developed and are used to compare tracking performance of the sequential and the parallel implementation of the algorithm. The non-simulation techniques are shown to accurately predict performance trends observed in simulations, that is the sequential implementation gives better tracking performance in terms of RMS position error and track lifetime. For the sequential implementation the question of processing order for different sensors is also briefly addressed. We also show that the sequential implementation is exponentially less computationally complex as the number of sensors increases. Thus, while sequential and parallel implementations are equivalent in multisensor filtering when no data association routine is needed, the sequential implementation gives superior tracking performance when data association is required.

Multisensor multitarget mixture reduction algorithms for tracking

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.