Statistical Analysis of Massive AIS Trajectories Using Gaussian Mixture Models

The Automatic Identification System (AIS) is an automatic tracking system which has been widely applied in the fields of intelligent transportation systems, e.g., collision avoidance, navigation, maritime supervision and management. Compare with other positioning systems, e.g., very high frequency (VHF) and radar, AIS can conquer the human errors and it is almost not affected by the external environment. To make better use of the AIS data, it is necessary to statistically analyze the massive AIS trajectories. The statistical results could make us better understand the potential properties of AIS trajectories. It is well known that most current practical applications are strongly dependent on the geometrical structures of AIS trajectories. In this paper, a Gaussian Mixture Model (GMM) is introduced to investigate the longitude and latitude differences of AIS trajectory data. The parameters of GMM are estimated using the Expectation Maximization (EM) algorithm. The experimental results have illustrated the superior performance of our proposed method.