Multiple Target Tracking Algorithms Research Articles

Tracking Multiple Vehicles Constrained to a Road Network From a UAV with Sparse Visual Measurements.

Multiple-target tracking algorithms generally operate in the local frame of the sensor and have difficulty with track reallocation when targets move in and out of the sensor field-of-view. This poses a problem when an unmanned aerial vehicle (UAV) is tracking multiple ground targets on a road network larger than its field-of-view. To address this problem, we propose a Rao-Blackwellized Particle Filter (RBPF) to maintain individual target tracks and to perform probabilistic data association when the targets are constrained to a road network. This is particularly useful when a target leaves and then re-enters the UAV’s field-of-view. The RBPF is structured as a particle filter of particle filters. The top level filter handles data association and each of its particles maintains a bank of particle filters to handle target tracking. The tracking particle filters incorporate both positive and negative information when a measurement is received. We implement two path planning controllers, receding horizon and deep reinforcement learning, and compare their ability to improve the certainty for multiple target location estimates. The controllers prioritize paths that reduce each target’s entropy. In addition, we develop an algorithm that computes the upper bound on the filter’s performance, thus facilitating an estimate of the number of UAVs needed to achieve a desired performance threshold.

Mean Shift Fusion Color Histogram Algorithm for Nonrigid Complex Target Tracking in Sports Video

We analyze and study the tracking of nonrigid complex targets of sports video based on mean shift fusion color histogram algorithm. A simple and controllable 3D template generation method based on monocular video sequences is constructed, which is used as a preprocessing stage of dynamic target 3D reconstruction algorithm to achieve the construction of templates for a variety of complex objects, such as human faces and human hands, broadening the use of the reconstruction method. This stage requires video sequences of rigid moving target objects or sets of target images taken from different angles as input. First, the standard rigid body method of Visuals is used to obtain the external camera parameters of the sequence frames as well as the sparse feature point reconstruction data, and the algorithm has high accuracy and robustness. Then, a dense depth map is computed for each input image frame by the Multi-View Stereo algorithm. The depth reconstruction with a too high resolution not only increases the processing time significantly but also generates more noise, so the resolution of the depth map is controlled by parameters. The multiple hypothesis target tracking algorithms are used to track multiple targets, while the chunking feature is used to solve the problem of mutual occlusion and adhesion between targets. After finishing the matching, the target and background models are updated online separately to ensure the validity of the target and background models. Our results of nonrigid complex target tracking by mean shift fusion color histogram algorithm for sports video improve the accuracy by about 8% compared to other studies. The proposed tracking method based on the mean shift algorithm and color histogram algorithm can not only estimate the position of the target effectively but also depict the shape of the target well, which solves the problem that the nonrigid targets in sports video have complicated shapes and are not easy to track. An example is given to demonstrate the effectiveness and adaptiveness of the applied method.

A GGIW-PHD Filter for Multiple Non-Ellipsoidal Extended Targets Tracking With Varying Number of Sub-Objects

When the extension state of the non-ellipsoidal extended target (NET) changes, the performance of traditional multiple target tracking algorithms based on the constant number of sub-objects will decrease. To solve this problem, this paper proposes a gamma Gaussian inverse Wishart probability hypothesis density filter for non-ellipsoidal extended targets with varying number of sub-objects, called VN-NET-GGIW-PHD filter. In the proposed filter, each NET is considered as a combination of multiple spatially close sub-objects, and the label management is introduced to realize the association between the NET and corresponding sub-objects. Then, by target spawning and combination, the number of sub-objects for approximating the extension state of each NET can be adjusted automatically. Furthermore, to obtain the partition of the measurement set, an approach based on the clustering by fast search and find of density peaks (CFSFDP) algorithm and expectation maximization (EM) algorithm is proposed. Simulation results show that the proposed filter can adaptively adjust the number of sub-objects and has better performance when the extension state of the NET changes.

Dwell Time Allocation Algorithm for Multiple Target Tracking in LPI Radar Network Based on Cooperative Game.

To solve the problem of dwell time management for multiple target tracking in Low Probability of Intercept (LPI) radar network, a Nash bargaining solution (NBS) dwell time allocation algorithm based on cooperative game theory is proposed. This algorithm can achieve the desired low interception performance by optimizing the allocation of the dwell time of each radar under the constraints of the given target detection performance, minimizing the total dwell time of radar network. By introducing two variables, dwell time and target allocation indicators, we decompose the dwell time and target allocation into two subproblems. Firstly, combining the Lagrange relaxation algorithm with the Newton iteration method, we derive the iterative formula for the dwell time of each radar. The dwell time allocation of the radars corresponding to each target is obtained. Secondly, we use the fixed Hungarian algorithm to determine the target allocation scheme based on the dwell time allocation results. Simulation results show that the proposed algorithm can effectively reduce the total dwell time of the radar network, and hence, improve the LPI performance.

Fast measurement partitioning algorithm for multiple extended target tracking

For tracking of multiple extended targets with unknown and varying target number, the measurements partitioning is indispensable and plays an important role in achieving good performance. However, the number of possible partitions grows very fast with the increasing of total number of measurements, which causes heavy computational burden. Thus this Letter proposes a fast measurement partitioning algorithm, in which the target predictive information is explicitly considered and used to determine the clustering centres for measurement partitioning. To guarantee its robustness, the target predictions are further modified when the target number is overestimated. This flexible strategy reduces the dependence on original predictive information and significantly improves the accuracy of partitioning results. Besides, the proposed algorithm greatly reduces the partition number by fully considering possible changes of the number of newborn targets and spawned targets. It not only eliminates unreasonable and redundant measurement partitions for tracking performance improvement, but also fits well with the tracking of newborn targets and spawned targets. Simulation results verify the benefit of what the authors proposed.

A message passing approach for multiple maneuvering target tracking

This paper considers the problem of detecting and tracking multiple maneuvering targets, which suffers from the intractable inference of high-dimensional latent variables that include target kinematic state, target visibility state, motion mode-model association, and data association. A unified message passing algorithm that combines belief propagation (BP) and mean-field (MF) approximation is proposed for simplifying the intractable inference. By assuming conjugate-exponential priors for target kinematic state, target visibility state, and motion mode-model association, the MF approximation decouples the joint inference of target kinematic state, target visibility state, motion mode-model association into individual low-dimensional inference, yielding simple message passing update equations. The BP is exploited to approximate the probabilities of data association events since it is compatible with hard constraints. Finally, the approximate posterior probability distributions are updated iteratively in a closed-loop manner, which is effective for dealing with the coupling issue between the estimations of target kinematic state and target visibility state and decisions on motion mode-model association and data association. The performance of the proposed algorithm is demonstrated by comparing with the well-known multiple maneuvering target tracking algorithms, including interacting multiple model joint probabilistic data association, interacting multiple model hypothesis-oriented multiple hypothesis tracker and multiple model generalized labeled multi-Bernoulli.

Loopy belief propagation based data association for extended target tracking

The data association problem of multiple extended target tracking is very challenging because each target may generate multiple measurements. Recently, the belief propagation based multiple target tracking algorithms with high efficiency have been a research focus. Different from the belief propagation based Extended Target tracking based on Belief Propagation (ET-BP) algorithm proposed in our previous work, a new graphical model formulation of data association for multiple extended target tracking is proposed in this paper. The proposed formulation can be solved by the Loopy Belief Propagation (LBP) algorithm. Furthermore, the simplified measurement set in the ET-BP algorithm is modified to improve tracking accuracy. Finally, experiment results show that the proposed algorithm has better performance than the ET-BP and joint probabilistic data association based on the simplified measurement set algorithms in terms of accuracy and efficiency. Additionally, the convergence of the proposed algorithm is verified in the simulations.

Joint Sensor Selection and Power Allocation Algorithm for Multiple-Target Tracking of Unmanned Cluster based on Fuzzy Logic Reasoning.

The unmanned aerial vehicle (UAV) cluster is gradually attracting more attention, which takes advantage over a traditional single manned platform. Because the size of the UAV platform limits the transmitting power of its own radar, how to reduce the transmitting power while meeting the detection accuracy is necessary. Aim at multiple-target tracking (MTT), a joint radar node selection and power allocation algorithm for radar networks is proposed. The algorithm first uses fuzzy logic reasoning (FLR) to obtain the priority of targets to radars, and designs a radar clustering algorithm based on the priority to form several subradar networks. The radar clustering algorithm simplifies the problem of multiple-radar tracking multiple-target into several problems of multiple-radar tracking a single target, which avoids complex calculations caused by multiple variables in the objective function of joint radar node selection and power allocation model. Considering the uncertainty of the target RCS in practice, the chance-constraint programming (CCP) is used to balance power resource and tracking accuracy. Through the joint radar node selection and power allocation algorithm, the radar networks can use less power resource to achieve a given tracking performance, which is more suitable for working on drone platforms. Finally, the simulation proves the effectiveness of the algorithm.

Autonomous collision detection and avoidance for ARAGON USV: Development and field tests

AbstractThis study addresses the development of algorithms for multiple target detection and tracking in the framework of sensor fusion and its application to autonomous navigation and collision avoidance systems for the unmanned surface vehicle (USV) Aragon. To provide autonomous navigation capabilities, various perception sensors such as radar, lidar, and cameras have been mounted on the USV platform and automatic ship detection algorithms are applied to the sensor measurements. The relative position information between the USV and nearby objects is obtained to estimate the motion of the target objects in a sensor‐level tracking filter. The estimated motion information from the individual tracking filters is then combined in a central‐level fusion tracker to achieve persistent and reliable target tracking performance. For automatic ship collision avoidance, the combined track data are used as obstacle information, and appropriate collision avoidance maneuvers are designed and executed in accordance with the international regulations for preventing collisions at sea (COLREGs). In this paper, the development processes of the vehicle platform and the autonomous navigation algorithms are described, and the results of field experiments are presented and discussed.

Improved Multiple Hypothesis Tracker for Joint Multiple Target Tracking and Feature Extraction

Feature-aided tracking can often yield improved tracking performance over the standard multiple target tracking (MTT) algorithms. However, in many applications, the feature signal of the targets consists of sparse Fourier-domain signals. It changes quickly and nonlinearly in the time domain, and the feature measurements are corrupted by missed detections and misassociations. In this paper, we develop a feature-aided multiple hypothesis tracker for joint MTT and feature extraction in dense target environments. We use the atomic norm constraint to formulate the sparsity of feature signal and use the ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -norm to formulate the sparsity of the corruption induced by misassociations. Based on the sparse representation, the feature signal are estimated by solving a semidefinite program. With the estimated feature signal, refiltering is performed to estimate the kinematic states of the targets, where the association makes use of both kinematic and feature information. Simulation results are presented to illustrate the performance of the proposed algorithm.

Fast density peak-based clustering algorithm for multiple extended target tracking

The key challenge of the extended target probability hypothesis density (ET-PHD) filter is to reduce the computational complexity by using a subset to approximate the full set of partitions. In this paper, the influence for the tracking results of different partitions is analyzed, and the form of the most informative partition is obtained. Then, a fast density peak-based clustering (FDPC) partitioning algorithm is applied to the measurement set partitioning. Since only one partition of the measurement set is used, the ET-PHD filter based on FDPC partitioning has lower computational complexity than the other ET-PHD filters. As FDPC partitioning is able to remove the spatially close clutter-generated measurements, the ET-PHD filter based on FDPC partitioning has good tracking performance in the scenario with more clutter-generated measurements. The simulation results show that the proposed algorithm can get the most informative partition and obviously reduce computational burden without losing tracking performance. As the number of clutter-generated measurements increased, the ET-PHD filter based on FDPC partitioning has better tracking performance than other ET-PHD filters. The FDPC algorithm will play an important role in the engineering realization of the multiple extended target tracking filter.

Bearings-only multi-target tracking using an improved labeled multi-Bernoulli filter

Most classical bearing-only target tracking algorithms model the measurement likelihood by one Gaussian distribution. The effectiveness of one Gaussian distribution model relies heavily on the accuracy of the predicted target position. However, due to the high nonlinearity of the bearing-only measurement, the predicted target position is mostly inaccurate before the target state observability is established. As a consequence, some classical nonlinear filters become not applicable for tracking bearing-only targets, especially when the measurements of multiple targets and clutter are present. The published bearings-only multiple-target tracking algorithms suffer from either the estimation inaccuracy or lack of track trajectories. Motivated by the problems mentioned above, we propose an improved labeled multi-Bernoulli filter for the goal of reducing estimation error under the premise that track trajectories are guaranteed. The proposed method divides the bearing measurement uncertainty into several measurement components that the measurement likelihood can be approximated by a Gaussian mixture. By assigning each track a unique label, the previous scan estimations and current scan measurements are associated and the track trajectories become available. Simulation results show that the proposed method considerably reduces estimation error. Further, various scenario parameters are investigated to validate the effectiveness of the proposed method.

Compressed Sensing-based Multiple-target Tracking Algorithm for Ad Hoc Camera Sensor Networks

Compressed Sensing-based Multiple-target Tracking Algorithm for Ad Hoc Camera Sensor Networks

Variational Bayesian adaptation of noise covariances in multiple target tracking problems

Multiple Target Tracking (MTT) is the process of computing the number of targets present in a surveillance area. MTT requires estimation of state variables and data association. New measurements are associated with existing tracks, clutter or new tracks. MTT generally involves unknown number of targets. Mostly because of computational complexity faced by MTT algorithms, it is a difficult and challenging problem. Computational load, underlying assumptions of known number of targets, and high cluttered environment are the main reasons, which available methods cannot address properly. Rao-Blackwellized has been used for multiple target tracking. It uses Kalman filter for state estimation and particle filter for data association. Our objective is to extend Rao-Blackwellized Monte Carlo Data Association (RBMCDA) that estimates number of targets and maintains track continuity enabling persistent tracking of targets. RBMCDA has been tested with seven different resampling methods in an effort to obtain the best resampling method. Gating validation and Variational Bayesian have been incorporated for multi target tracking problem. The modified RBMCDAs are applied to different case studies for its performance evaluation.

Iterative joint integrated probabilistic data association filter for multiple-detection multiple-target tracking

Most classical target tracking algorithms assume that one target generates one measurement per scan. However, for situations where the target's physical size exceeds one sensor resolution cell or the transmission and received signals have multi-path propagations, one target generates multiple detections which terms the multiple-detection problem. The multiple-detection multiple-target tracking algorithms usually suffered from an intractable computational load and could not operate in real time when multiple targets are closely spaced or too many detections are generated. In addition, automatic track initialization technique in cluttered environments brings in the presences of both true and false tracks that false track discrimination is required. These two terms are critical for real tracking systems but largely neglected by the published papers for the multiple-detection problem. In this paper, the authors propose an algorithm, called the Multiple-detection Iterative Joint Integrated Probabilistic Data Association (MD-iJIPDA) filter, that iteratively adjusts the computational load as well the tracking performances so that desirable performances can be achieved within manageable time. By integrating the target existence probability as a track quality measure, the MD-iJIPDA becomes capable of false track discrimination. The MD-iJIPDA is obtained by utilizing the arithmetic structure of MD-JIPDA algorithm. Therefore, we first incorporate the target existence probability into the calculation of joint event probability in MD-JPDA to derive the MD-JIPDA algorithm, then followed by the derivation of MD-iJIPDA algorithm. Simulations for the scenarios of large number targets with multiple detections are investigated to validate the effectivenesses of the proposed algorithm in tracking performance and computational efficiency.

Tracking multiple moving objects in images using Markov Chain Monte Carlo

A new Bayesian state and parameter learning algorithm for multiple target tracking (MTT) models with image observations is proposed. Specifically, a Markov chain Monte Carlo algorithm is designed to sample from the posterior distribution of the unknown number of targets, their birth and death times, states and model parameters, which constitutes the complete solution to the tracking problem. The conventional approach is to pre-process the images to extract point observations and then perform tracking. We model the image generation process directly to avoid potential loss of information when extracting point observations. Numerical examples show that our algorithm has improved tracking performance over commonly used techniques, for both synthetic examples and real florescent microscopy data, especially in the case of dim targets with overlapping illuminated regions.

Online Scheme for Multiple Camera Multiple Target Tracking Based on Multiple Hypothesis Tracking

We propose an online tracking algorithm for multiple target tracking with multiple cameras. In this paper, we suggest a multiple hypothesis tracking (MHT) framework to find an unknown number of multiple tracks through the spatio-temporal association between tracklets generated from multiple cameras. In this framework, the MHT is realized online by solving the maximum weighted clique problem (MWCP) at every frame to estimate the 3D trajectories of the targets. To handle the NP-hard issue of the MWCP, we propose a novel online scheme that formulates the MWCP using feedback information from the previous frame’s result to find optimal tracks at every frame. This scheme enables the MWCP to be formulated by multiple subproblems and will significantly reduce the computation. The experiments show that the proposed algorithm performs comparably with the state-of-the-art batch algorithms, even though it adopts an online scheme.

Probabilistic 3D Sound Source Mapping System Based on Monte Carlo Localization Using Microphone Array and LIDAR

[abstFig src='/00290001/09.jpg' width='300' text='3D sound source environmental map' ] The study proposes a probabilistic 3D sound source mapping system for a moving sensor unit. A microphone array is used for sound source localization and tracking based on the multiple signal classification (MUSIC) algorithm and a multiple-target tracking algorithm. Laser imaging detection and ranging (LIDAR) is used to generate a 3D geometric map and estimate the location of its six-degrees-of-freedom (6 DoF) using the state-of-the-art gyro-integrated iterative closest point simultaneous localization and mapping (G-ICP SLAM) method. Combining these modules provides sound detection in 3D global space for a moving robot. The sound position is then estimated using Monte Carlo localization from the time series of a tracked sound stream. The results of experiments using the hand-held sensor unit indicate that the method is effective for arbitrary motions of the sensor unit in environments with multiple sound sources.

Comparison and Analysis of Recursive-RANSAC for Multiple Target Tracking

In both simulated and real experiments, the recursive random sample consensus (R-RANSAC) algorithm has shown promise as an efficient multiple target tracking (MTT) filter. In this paper, we introduce a generalized and modular framework for the R-RANSAC algorithm, analyze the sensitivity of the R-RANSAC tuning parameters, and compare R-RANSAC to five well-known MTT algorithms. We show that R-RANSAC offers a unique balance between low computational complexity, excellent track continuity, and good performance in cluttered environments.

Soft and evolutionary computation based data association approaches for tracking multiple targets in the presence of ECM

This paper proposes two novel soft and evolutionary computing based hybrid data association techniques to track multiple targets in the presence of electronic countermeasures (ECM), clutter and false alarms. Joint probabilistic data association (JPDA) approach is generally used for tracking multiple targets. Fuzzy clustering means (FCM) technique was proposed earlier as an efficient method for data association, but its cluster centers may fall to local minima. Hence, new hybrid data association approaches based on fuzzy particle swarm optimization (Fuzzy-PSO) and fuzzy genetic algorithm (Fuzzy-GA) clustering techniques have been presented as robust methods to overcome local minima problem. The data association matrix is evaluated for all tracks using validated measurements obtained by phased array radar for four different cases applying four data association methods (JPDA, FCM, Fuzzy-PSO, and Fuzzy-GA). Therefore, two hybrid data association approaches are designed and tested for multi-target tracking using intelligent techniques. Experimental results indicate that Fuzzy-GA data association technique provides improved performance compared to all other methods in terms of position and velocity RMSE values (38.69% and 33.19% average improvement for target-1;31.17% and 9.68% average improvement for target-2) respectively for crossing linear targets case. However, FCM technique gives better performance in terms of execution time (94.88% less average execution time) in comparison with other three techniques(JPDA, Fuzzy-GA, and Fuzzy-PSO) for the case of linear crossing targets. Thus accomplishing efficient and alternative multiple target tracking algorithms based on expert systems. The results have been validated with 100 Monte Carlo runs.