Enhanced Target Ship Tracking With Geometric Parameter Estimation for Unmanned Surface Vehicles

Abstract

Autonomous collision detection and avoidance is a crucial requirement for the safe navigation of unmanned surface vehicles (USVs) in maritime traffic situations. Automatic identification system (AIS) is used to obtain the motion information of surrounding ships and their dimensional specifications. With AIS information, appropriate collision risk assessment between two ships can be performed; further, collision avoidance can be achieved by defining a safe radius of avoidance, which can be determined considering the shape parameters of a target ship. However, AIS data are often unreliable and some commercial fishing vessels intentionally turn off the public tracking system to hide their location. Under these circumstances, marine radars are used to detect and estimate the motion information of nearby ships. However, most existing target tracking studies model the target as a point object without any spatial extent, and thus, its physical dimensions cannot be identified. In this paper, a target tracking method that uses a marine radar is proposed to simultaneously estimate the motion states (i.e., position, course, and speed) of a target ship and its geometric parameter (length) in the framework of an extended Kalman filter (EKF). The proposed approach enhances collision avoidance by providing the kinematic and length parameters to evaluate collision risk and generate an appropriate collision-free path. Real-sea experiments using a developed USV system were conducted to verify and demonstrate the feasibility of the proposed algorithm; the results are presented and discussed in this paper.

Robust Decision-Making for the Reactive Collision Avoidance of Autonomous Ships against Various Perception Sensor Noise Levels

“…Han et al [40] used an extended Kalman filter (EKF) for electrooptical and infrared camera, RADAR, and LIDAR measurements to predict the motion of obstacles. Han et al [41] used EKF on RADAR measurements to predict the motion of obstacles. Stanislas et al and Kim et al [42] filtered RADAR and LIDAR measurements by experimentally tuning the intensity of basic RADAR settings and filtering out LIDAR measurements that exceeded a predefined threshold to estimate the state of obstacles.…”

An Overview of the PAKF-JPDA Approach for Elliptical Multiple Extended Target Tracking Using High-Resolution Marine Radar Data

“…Detailed reviews on extent approximation methods can be found in [5,14,15]. Ellipses have so far been the most favored basic shape to approximate vessels' extents in maritime applications, where we represent the length and width by means of the ellipse's axes and the true heading by the ellipse's orientation [16][17][18][19]. Traditional methods for elliptical target tracking are based on the Random Matrix Method (RMM) framework, where a target's kinematic state is modeled as a Gaussian distribution and its extent by a Wishart distribution, which is represented using an Symmetric and Positive Definite (SPD) matrix [6,[20][21][22].…”