Research on Submarine Localization Based on K-means Clustering Algorithm

With the development of submarines and the increasing concern about the safety of this program, this study constructs a localization model and a search model for search and rescue after an accident. In addition, we provide a series of equipment recommendations and fundamentals to minimize the risk of accidents. We deeply analyze the influencing factors and obtain the key factors (ocean currents, seawater density, and seafloor topography), and then we process the obtained data related to the key factors, such as clustering, rasterization, and visualization. Using the processed data, we propose a three-dimensional physical model to analyze the impact on the submersible. Based on this physical model, we constructed a multilayer cross-location model (MLCL) to predict the positional trajectory of the submersible over time. To verify the accuracy of the model predictions, we propose a probability distribution model to comparatively validate the MLCL model. The final results show that the accuracy of the MLCL model is above 91%.