Applied Sciences (Jan 2020)

Vessel Trajectory Reconstruction Based on Functional Data Analysis Using Automatic Identification System Data

  • Myeong-Hun Jeong,
  • Seung-Bae Jeon,
  • Tae-Young Lee,
  • Min Kyo Youm,
  • Dong-Ha Lee

DOI
https://doi.org/10.3390/app10030881
Journal volume & issue
Vol. 10, no. 3
p. 881

Abstract

Read online

This study provides an automatic shipping-route construction method using functional data analysis (FDA), which analyzes information about curves, such as multiple data points over time. The proposed approach includes two steps: outlier detection and shipping-route construction. This study uses automatic-identification system (AIS) data for the experiments. The effectiveness of the proposed method is demonstrated through case studies, wherein our approach is compared with the Mahalanobis distance method for trajectory-outlier detection, and the performance of vessel trajectory reconstruction is compared with that of a density-based approach. The proposed method improves understanding of vessel-movement dynamics, thereby improving maritime monitoring and security.

Keywords