Vessel Trajectory Reconstruction Of Missing AIS Data Values Machine Learning.


The report discusses the importance of Automatic Identification System (AIS) data in tracking vessel trajectories and its applications in various fields such as mapping shipping density, characterizing marine traffic patterns, anomalous behavior detection of ships, collision risk analysis, investigation of maritime accidents, and maritime route generation for vessels. The report also highlights the challenges of working with AIS data, which is often plagued with inconsistencies and noise. Additionally, it provides an overview of several works on trajectory quality improvement, including methods such as Piecewise Linear Interpolation, Piecewise Cubic Interpolation, cubic Hermit interpolation, and the discrete Kalman algorithm. The paper concludes by suggesting future research directions in the field of AIS data analysis.

Key highlights
  • AIS is an automatic vessel tracking system that broadcasts important messages describing the vessel and its sailing status information
  • Ship trajectories have various critical applications, including mapping shipping density, characterizing marine traffic patterns, detecting anomalous behaviour of ships, collision risk analysis, investigating maritime accidents, and generating maritime routes for vessels.
  • Notable works have been done for each application, such as visualizing ship routes, deriving shipping density maps, detecting and mapping fishing activities, and developing anomaly detectors for detecting anomalous vessel behaviour.
  • Trajectory quality improvement has been addressed by various approaches such as Piecewise Linear/Cubic/Spline Interpolation, trajectory restoration based on navigational features of the vessel, linear interpolation, cubic Hermit interpolation, and discrete Kalman algorithm.
Key Challenges
  • The accuracy and precision of AIS data can be affected by various factors, such as the quality of the AIS device, signal interference, and the vessel’s speed and direction.
  • AIS data contains sensitive information about vessels and their activities, which can raise privacy and security concerns.
  • AIS data comes from various sources and is stored in different formats, making it challenging to standardize and integrate into a unified system.
  • Processing and analysing large amounts of AIS data can be computationally intensive and time-consuming, requiring advanced algorithms and techniques.
Major Opportunities
  • By using AIS data to construct ship trajectories and analyse vessel behaviour, it is possible to identify anomalous behaviour that could indicate a potential safety risk or collision risk. This information can be used to prevent accidents and improve safety measures.
  • AIS data can be used to map shipping density and traffic patterns, which can help in the planning of maritime activities and infrastructure. This can lead to better utilization of resources and more efficient transportation routes.
  • By employing data mining and cleaning methods, AIS data can be made more reliable and useful for various applications. This can lead to the development of new algorithms, models, and tools that can improve the analysis and utilization of AIS data.

“AIS data can enhance maritime safety and efficiency, and also the ongoing research and development efforts to improve its quality and usability.”

Dhanush Balaji R D, Dr. (Cmdr) Arnab Das