High-precision intelligent track prediction under multi-dimensional conditions

In order to maintain national maritime security, it is particularly important to master the navigation rules of ships on important maritime military routes. Traditional trajectory prediction methods cannot flexibly implement the function of autonomous re-prediction based on emergencies. This paper considers the large amount of trajectory data of maritime moving targets and the existence of multi-dimensional factors and presents a high-precision track prediction model under multi-dimensional factors. This algorithm uses the Douglas-Peucker algorithm to compress the track data and uses the DBSCAN(Density-Based Spatial Clustering of Applications with Noise) algorithm for trajectory clustering and predicts the trajectory through the LRCN(Long-term Recurrent Convolutional Networks) prediction model. Simulation results show that using this algorithm greatly reduces the amount of calculations and can achieve a high-precision prediction; Under emergency cases, the algorithm can achieve autonomous re-prediction with high accuracy.