Unmanned Surface Vessel (USV) can be equipped with photoelectric pods and marine radars to detect the ship targets of sea surface to obtain the type and location of the ships. Aiming at the characteristics of data attributes and dimensions in the fusion of two sensors, the data fusion algorithm based on the Grey Relation Analysis (GRA) is proposed. This method first carried out sensors registration for the radar and photoelectric pod, and then extracted the characteristics of the horned trace characteristics of the target in different sensors. The similarity of the changing trend of the angle trace curve through GRA was judged. Finally, the information such as the target category, the track, the latitude and longitude, and the distance from the USV were obtained. The results show that the data fusion algorithm based on GRA can complete the target attribute correlation, and can meet the needs of the environmental sensing task for USV.