1,721,042 research outputs found
Joint probabilistic data association tracker for extended target tracking applied to X-band marine radar data
No full textX-band marine radar systems are flexible and low-cost tools for monitoring multiple targets in a surveillance area. Although they may suffer from several sources of interference, e.g., sea clutter, they can provide high-resolution measurements in both space and time. Such features offer the opportunity to get accurate information not only about the target kinematics, i.e., positions and velocities, as other conventional radars, but also about the targets' extents. This research area is named extended target tracking (ETT). In this paper, we propose a signal processing chain composed by a detector and a joint probabilistic data association (JPDA) tracker to handle the problem of multiple ETT and to jointly estimate both the targets' kinematics and their sizes, i.e., length and width. The performance assessment is conducted on real data acquired by an X-band marine radar located in the Gulf of La Spezia, Italy. The experimental results demonstrate the ability of the processing chain to reach high performance with a limited computational burden
Extended target tracking using joint probabilistic data association filter on X-band radar data
No full textX-band Marine radar systems are low-cost tools for monitoring multiple targets in a surveillance area. Although they may suffer from several sources of interference, high resolution measurements in both space and time can be provided. Such features offer the opportunity to get accurate information not only about the targets' kinematics, as other conventional sensors, but also about the targets' extent. In this paper, a signal processing chain composed by a detector and a joint probabilistic data association tracker is proposed to address the problem of tracking using X-band Marine radar data. Estimations of both the targets' kinematics, i.e. positions and velocities, and length and width, are provided. The performance assessment, conducted on real data acquired by an X-band Marine radar located in the Gulf of La Spezia, Italy, demonstrates the ability of the processing chain to obtain high tracking performance with a limited computational burden
Maximum likelihood estimation in a parametric stochastic trajectory model
No full textIn this work, we consider maximum likelihood estimation of parameters in a stochastic trajectory model. The velocity paths are generated from an Ornstein-Uhlenbeck process and thus revert to a latent expected value. In addition to this expected velocity, parameters that specify the reversion characteristics and the process noise covariance determine the behaviour of typical trajectories of the model. Estimation of these parameters from trajectory samples facilitates learning of patterns and training of predictive models using trajectory data, e.g., automatic identification system (AIS) messages transmitted by vessels. We propose a six-degrees-of-freedom parameterisation and investigate the identifiability of these parameters using the Cramér-Rao bound matrix which we estimate using Monte Carlo methods. We demonstrate that some parameter configurations of interest are identifiable and their maximum likelihood estimate can be found using iterative optimisation algorithms. We demonstrate the efficacy of this approach on both simulated and real data
Data Driven Vessel Trajectory Forecasting Using Stochastic Generative Models
No full textIn this work, we propose a data driven trajectory forecasting algorithm that utilizes both recorded historical and streaming trajectory observations. The algorithm performs Bayesian inference on a directed graph the walks on which represent stochastic change point models of trajectory classes. Parameter distributions of these models are learnt from recorded trajectories. Forecasting is then made by calculating the class - or, walk- probabilities and corresponding predictive distributions for a given stream of location and velocity observations. This approach is tailored for the maritime domain and automatic identification system (AIS) data exploitation through the use of an Ornstein-Uhlenbeck process driven stochastic process model that captures vessel motion characteristics. We demonstrate the efficacy of this approach on a real data set
Knowledge-based ship tracking applied to HF surface wave radar data
No full textIn recent years, low-power high-frequency surface-wave radars have received significant attention thanks to their over-the-horizon coverage capability and the continuous-time operation mode. These radars have become effective long-range early-warning tools for maritime situational awareness applications. In this paper a knowledge-based multi-target tracking algorithm is described. The advantages in using a prior information on ship traffic are assessed exploiting real data acquired by two high-frequency surface-wave radars. The outcomes confirm the ability of the proposed approach to better follow targets with a time-on-target increment up to 30% with respect to existing methods. A reduction of the track fragmentation up to 20% is also observed
Variable structure interacting multiple model algorithm for ship tracking using HF surface wave radar data
No full textThese last decades spawned a great interest towards low-power High-Frequency (HF) Surface-Wave (SW) radars for ocean remote sensing. By virtue of their over-the-horizon coverage capability and continuous-time mode of operation, these sensors are also effective long-range early-warning tools in maritime situational awareness applications. In this paper we show how it is possible to take advantage of a priori information on traffic by the means of a knowledge-based multi-target tracking algorithm, demonstrating that the tracking stage can be enhanced by combining on-line data from the HFSW radar with ship traffic information. A significant improvement of the proposed procedure, in terms of system performance, is demonstrated in comparison with the state-of-the-art approach recently presented in the literature. The main benefit of our approach is the ability to better follow targets without increasing the false alarm rate. The ability to follow targets can be over 30% better than existing methods. The proposed approach also exhibits a reduction of the track fragmentation. Average gains between the 13% and the 20% are observed
Knowledge-Based Multitarget Ship Tracking for HF Surface Wave Radar Systems
No full textThese last decades spawned a great interest toward low-power high-frequency (HF) surface-wave (SW) radars for ocean remote sensing. By virtue of their over-the-horizon coverage capability and continuous-time mode of operation, these sensors are also effective long-range early warning tools in maritime situational awareness applications providing an additional source of information for target detection and tracking. Unfortunately, they also exhibit many shortcomings that need to be taken into account, and proper algorithms need to be exploited to overcome their limitations. In this paper, we develop a knowledge-based (KB) multitarget tracking methodology that takes advantage of a priori information on the ship traffic. This a priori information is given by the ship sea lanes and by their related motion models, which together constitute the basic building blocks of a variable structure interactive multiple model procedure. False alarms and missed detections are dealt with using a joint probabilistic data association rule and nonlinearities are handled by means of the unscented Kalman filter. The KB-tracking procedure is validated using real data acquired during an HF-radar experiment in the Ligurian Sea (Mediterranean Sea). Two HFSW radar systems were operated to develop and test target detection and tracking algorithms. The overall performance is defined in terms of time-on-target, false-alarm rate (FAR), track fragmentation (TF), and accuracy. A full statistical characterization is provided using one month of data. A significant improvement of the KB-tracking procedure, in terms of system performance, is demonstrated in comparison with a standard joint probabilistic data association tracker recently proposed in the literature to track HFSW radar data. The main improvement of our approach is the better capability of following targets without increasing the FAR. This increment is much more evident in the region of low FAR, where it can be over the 30% for both the HFSW radar systems. The KB-tracking exhibits on average a reduction of the TF of about the 20% and the 13% of the utilized HFSW-radar systems
Extended target tracking applied to X-band marine radar data
No full textX-band marine radar systems are flexible and low-cost tools for monitoring multiple targets in a surveillance area. They are able to provide high resolution measurements in both space and time. Such features offer the opportunity to get accurate information not only about the targets' kinematics but also about the targets' extents. The tracking of these kinds of data is usually called extended target tracking (ETT). In this paper, we propose a signal processing chain mainly composed by a pixel-wise detector and a joint probabilistic data association tracker to handle the problem of multiple ETT. The performance assessment is conducted on real data acquired by an X-band marine radar located in the Gulf of La Spezia, Italy. The experimental results demonstrate that the processing chain is able to reach high performance with a limited computational burden
Consistent Estimation of Randomly Sampled OrnsteinUhlenbeck Process Long-Run Mean for Long-Term Target State Prediction
No full textIn this letter, we study the problem of estimating the long-run mean of the Ornstein-Uhlenbeck (OU) stochastic process and its effect on the long-term prediction of future vessel states, which is a crucial problem for Maritime Situational Awareness (MSA). We employ a sample mean estimator (SME) to estimate the key OU parameter from the observations, computing the closed-form SME covariance error in both the random and constant sampling time regimes, providing a fundamental building block of the overall long-term state prediction covariance. We show also that the SME is: root n-consistent when the sampling time is random; asymptotically efficient when the sampling time is constant; and very close to the Cramer-Raolower bound in the cases of practical interest for MSA
Unsupervised extraction of maritime patterns of life from Automatic Identification System data
No full textThis paper presents an unsupervised approach to extract maritime Patterns of Life (PoL) from historical Automatic Identification System (AIS) data based on a low-dimensional synthetic representation of ship routes. Recent advances in long-term vessel motion modeling through Ornstein-Uhlenbeck mean-reverting stochastic processes allow to encode knowledge about maritime traffic via a compact graph-based model where waypoints are graph vertices and the connections between them, i.e., the navigational legs, are graph edges. The resulting directed graph ultimately leads to the detection and statistical characterization of recurrent maritime traffic patterns. To demonstrate its effectiveness and applicability to real-world case studies, the proposed methodology has been tested on two extensive AIS datasets, collected in the areas of two operational trials of EU-H2020's MARISA (Maritime Integrated Surveillance Awareness) project
- …
