Dolphin 1 Ship Trajectory Prediction based on Ivy-multi-output Least Squares Support Vector Regression

Trajectory prediction for ships plays a crucial role in intelligent ship route planning and collision avoidance. The Dolphin 1 ship is the first digital twin intelligent ship of China and is developed by Harbin engineering university. To design a trajectory prediction model for the Dolphin 1 ship, this paper employs the multi-output least squares support vector regression (MLSSVR) method. The Ivy algorithm is also used to find the optimal hyper-parameters for MLSSVR. The performance of the established trajectory prediction model is validated using real-measured trajectory data. The experimental results demonstrate the generalizability and prediction accuracy of MLSSVR.