Fusion Detection Algorithm of Maritime Radar and Electro-Optical Pod for Complex Sea Conditions

In order to improve the perceptual capability of Unmanned Surface Vessels (USVs) in complex sea conditions, a fusion detection algorithm has been developed that combines maritime radar and electro-optical (EO) pod. This algorithm integrates the positional state of the USV, class information of the maritime target obtained through a vision-based target detection algorithm, and processed maritime radar information. By using Gaussian and morphological filtering, the fusion of position state information from the USV, EO pod data, and maritime radar data is achieved. This effectively mitigates the negative impact of environmental factors, such as waves and water mist, on target detection. A cooperative sensing process between the maritime radar and EO pod is proposed to fuse the data obtained from both sources, resulting in accurate comprehensive target information. Experimental results show that the fusion algorithm achieves an average accuracy of 80.24% and a detection rate of 74.7%, enabling precise detection of maritime targets, including their distance, speed, and category. This enhanced perceptual capability is essential for USVs operating in complex maritime environments, enabling better decision-making and navigation performance.