Ship track prediction based on Bayesian optimization in time convolutional networks

Abstract: Objective In order to improve the accuracy and calculation efficiency of ship track prediction, the traditional method is prone to gradient explosion and long calculation time. Method A track prediction model based on improved Bayesian optimization algorithm (IBOA) and time convolution neural network (TCN) was proposed. The time pattern attention mechanism (TPA) is introduced to extract the weights of each input feature and ensure the timing of historical data of flight track. At the same time, the reversible residual network is introduced to reduce the memory occupied by TCN model training. Finally, the superparameters (kernel size K, expansion coefficient d) in the time convolutional network were optimized by using Bayesian optimization algorithm, and the optimal model was obtained for track prediction. Result The track data collected by AIS was verified, and the root mean square error (RMSE) was increased by 5.5×10-5, 3.5×10-4 and 6×10-4 in weak coupling,