Track Predition Based on CNN-GRU Neural Network

Automatic identification system (AIS) data, which might mirror the characteristics of the ship's navigation standing in real time, and provides ship's characteristics for ship's track predition, that is of great significance to avoid water traffic accidents. so as to enhance the accuracy and stability of ship track prediction, this paper proposes a combined deep learning network supported AIS ship navigation information training, and predicts the track supported convolutional neural network (CNN) gate recurrent unit (GRU). The longitude and latitude, heading and speed of the ship's AIS information are used as input data, and the longitude and latitude are used as output data to build a prediction model. The results show that the MSE of this prediction model reaches 7.786e-05, which might accurately predict the ship's track.