Real-Time Underwater Object Detection Based on DC Resistivity Method

Abstract

A new real-time underwater object detection method adopting geophysical direct-current resistivity techniques is proposed. Our final goals are real-time detection and tracking of small submarines in water depths less than 100 m, under acoustically noisy conditions. The main features of our method are as follows: 1) a detection line, several kilometers long and semipermanently buried in the seabed, consisting of two current electrodes and multiple potential electrodes; 2) fixed current electrode configurations for high-speed real-time detection of the target object; 3) measurement of extremely low-level electric field signals, made possible by the shielding effect of the conductive seawater overburden; and 4) signals from a moving object such as a submarine are extracted by differential data analysis because the target object passes through the detection area for only a short time compared with the long-term drift change of the electrical properties of seawater. We verify our method and confirm real-time detection feasibility through numerical and physical scale experiments. For this purpose, a dedicated instrument and a background data update algorithm are developed, and a differential data analysis method is used. The background data update algorithm is used for counteracting the time-varying electrical characteristics of seawater. The differential data analysis method is exploited to extract disturbed signals by objects from complex background data. Through numerical experiments, we find that the $x$ -coordinate value of a target object corresponds to the position of a peak data point, and this result is consistent with 1:200 downscaled water tank experiments. We also confirm real-time detection feasibility using our instrument through an offshore experiment. We expect that our proposed method will complement conventional detection methods for harbor defense and surveillance systems in acoustically noisy environments.