Real-Time Localization for Underwater Moving Object Using Precalculated DC Electric Field Template

Abstract

Underwater object detection and tracking are of considerable interest in offshore defense. Recently, a new method that takes advantage of direct-current resistivity survey techniques has been introduced as a complement to the conventional detection method for use in acoustically noisy conditions. However, dedicated rapid target localization methods are required to enable real-time tracking in this detection method. Therefore, in this paper, we propose a real-time localization method using grid-based template matching. The main features of our method are as follows: 1) measurement data sets from two separate detection lines are used; 2) template matching based on correlation is used; 3) the templates are precalculated by numerical modeling; and 4) real-time localization processing is implemented by efficient calculations. To verify our method, we first performed experiments for stationary target object positions under a water tank condition: template versus numerical modeling data and template versus physical modeling data. Next, real-time localization experiments for a moving target in the water tank were conducted using the real-time data at 3-Hz refresh rate. In the real-time experiments, we confirmed that the estimated positions could keep track of the actual target positions continuously. We expect that this paper will contribute positively to offshore defense and surveillance.