A Performance Evaluation Scheme for Multiple Object Tracking with HFSWR.

Kun Wang,Weifeng Sun,Jiong Niu,Lun Zhao,Pengju Zhang,Yonggang Ji

doi:10.3390/s19061393

Abstract

High-frequency surface wave radar (HFSWR) can detect and continuously track ship objects in real time and beyond the horizon. When ships navigate in a sea area, their motions in a time period form a scenario. The diversity and complexity of the motion scenarios make it difficult to accurately track ships, in which failures such as track fragmentation (TF) are frequently observed. However, it is still unclear how and to what degrees the motions of ships affect the tracking performance, especially which motion patterns can cause tracking failures. This paper addresses this problem and attempts to undertake a first step towards providing an intensive quantitative performance assessment and vulnerability detection scheme for ship-tracking algorithms by proposing an evolutionary and data-mining-based approach. Low-dimensional scenarios in terms of multiple maneuvering ship objects are generated using a grammar-based model. Closed-loop feedback is introduced using evolutionary computation to efficiently collect scenarios that cause more and more tracking performance loss, which provides diversified cases for analysing using data-mining technique to discover indicators of tracking vulnerability. Results on different tracking algorithms show that more cluster and convergence patterns and longer duration of our convoy and cluster patterns in the scenarios can cause severer TF to HFSWR ship tracking.

Highlights

High-frequency surface wave radar (HFSWR) has been investigated as a platform for detection and tracking of moving ships in surveillance systems due to its advantage of all-day operation with low cost and wide observable region
We present the results we obtained to show the effectiveness of our proposed evolutionary and data-mining-based approach in collecting high-fidelity testing datasets with different degrees of tracking performance characteristics, achieving intensive and quantitative performance assessment of a HFSWR ship tracking algorithm, and in discovering indicators of tracking vulnerability by analysing these datasets
Performance assessment and vulnerability detection was implemented on two types of tracking algorithms—the converted measurement Kalman filter [22,23,24] and the Gaussian mixture probability hypothesis density filter (GM-PHD) [25,26]

Summary

Introduction

High-frequency surface wave radar (HFSWR) has been investigated as a platform for detection and tracking of moving ships in surveillance systems due to its advantage of all-day operation with low cost and wide observable region. HFSWR systems exhibit many shortcomings that degrade the detection performance, such as poor range and azimuth resolution, high nonlinearity in the state/measurement space, and significant false alarm rate (FAR) due to both sea clutter and man-made/natural interference [1]. To improve the accuracy of ship objects’ direction of arrival (DOA) estimation, beam forming and direction finding techniques have been applied [3,4]. Their performances are “easy to be affected by array uncertainties such as sensor position error and gain/phase errors” [4], which may introduce large errors into DOA estimations of up to Sensors 2019, 19, 1393; doi:10.3390/s19061393 www.mdpi.com/journal/sensors

Results

Discussion

Conclusion