Multisensor Feature Selection for Maritime Target Estimation

This paper introduces a preprocessing and feature selection technique for maritime target estimation. Given the distinct challenges of the maritime environment and the use of multiple sensors, we propose a target estimation model designed to achieve high accuracy while minimizing computational costs through suitable data preprocessing and feature selection. The experimental results demonstrate excellent performance, with the mean square error (MSE) reduced by about 99%. This approach is expected to enhance vessel tracking in situations where vessel estimation sensors, such as the automatic identification system (AIS), are disabled. By enabling reliable vessel tracking, this technique can aid in the detection of illegal vessels.