Back to top

Maritime Situational Awareness

Maritime
Trajectory
Maritime Situational Awareness

According to the International Maritime Organization IMO, 90% of the global trade is handled by the shipping industry. Maritime monitoring systems support safe shipping by detecting, in real-time, dangerous, suspicious and illegal vessel activities. Such systems typically use the Automatic Identification System AIS, a tracking technology for locating vessels at sea through data exchange. We have been developing a complex event recognition system for maritime monitoring in the Event Calculus that uses AIS as its main source of information, along with contextual geographical information (e.g, Fishing areas, Natura 2000 areas, depth information etc.).

alt text

Recognition & Forecasting methods

We perform complex event recognition using the ‘Event Calculus for Run-Time Reasoning’ (RTEC), an open-source Prolog implementation of the Event Calculus, designed to compute continuous temporal projection queries for pattern matching on data streams. Furthermore, we perform complex event forecasting using Wayeb, an open-source Scala implementation of symbolic automata and prediction suffix trees.

Detected and forecast maritime activities

We detect/forecast:

  • anchored vessels,
  • drifting vessels (true heading deviates from the course over ground),
  • tugging operations,
  • piloting,
  • trawling,
  • speed violations,
  • vessel rendezvous,
  • loitering and,
  • search and rescue operations.

See the videos below for some examples (best viewed in full screen mode):


Datasets

The detected instances of all composite maritime activities mentioned above, over a period of 6 months (October 2015-March 2016), in the area of Brest, France, is available here. The AIS signals of the dataset are available here, while a semantic annotation of the dataset is available here.


Publications

  • Fikioris G., Patroumpas K., Artikis A., Pitsikalis M., Paliouras G., Optimizing Vessel Trajectory Compression for Maritime Situational Awareness.
    In Geoinformatica, To Appear. PDF DOI

  • Katzouris N., Paliouras G. and Artikis A., Online Learning Probabilistic Event Calculus Theories in Answer Set Programming.
    In Theory and Practice of Logic Programming , 23(2), 362–386, 2023. PDF BibTeX DOI

  • Tsilionis E., Artikis A., Paliouras G., Incremental Event Calculus for Run-Time Reasoning.
    In Journal of Artificial Intelligence Research (JAIR), 73, pp. 967–1023, 2022. PDF BibTeX code DOI

  • Alevizos E., Artikis A., Paliouras G., Complex Event Forecasting with Prediction Suffix Trees.
    In The International Journal on Very Large DataBases (VLDBJ), 31(1):157–180, 2022. PDF BibTeX DOI

  • Fikioris G., Patroumpas K., Artikis A., Paliouras G. and Pitsikalis M., Fine-Tuned Compressed Representations of Vessel Trajectories.
    In International Conference on Information and Knowledge Management (CIKM), 2020. PDF Slides BibTeX code DOI

  • Pitsikalis M., Artikis A., Dreo R., Ray C., Camossi E., and Jousselme A., Composite Event Recognition for Maritime Monitoring.
    In 13th International Conference on Distributed and Event-Based Systems (DEBS), pp. 163–174, 2019 PDF Slides BibTeX DOI

  • Patroumpas K., Alevizos E., Artikis A., Vodas M., Pelekis N., Theodoridis Y., Online event recognition from moving vessel trajectories.
    In GeoInformatica, 21(2), pp. 389–427, 2017. PDF BibTeX DOI


Funding