In maritime surveillance, the detection, identification and management of inconsistencies in available information is of crucial importance for understanding the situation and analyzing ship behaviors. These inconsistencies may reflect illicit or malicious acts, ships in distress, or simply sensor malfunction. This task of monitoring is particularly difficult for an operator because of the volume, the variety and the lack of veracity of information. Moreover, the characterization of what constitutes an inconsistency represents a challenge in itself since it depends strongly on the operational context (mission, weather, geopolitics, environment).
The objective of the project is to develop a formal and tooled approach to help reasoning in the presence of inconsistencies. Building on previous work [1-5] including recent work at LGI2A and CMRE [6-10], the project will aim at proposing one or more measures of the degree of inconsistency having clear semantics and behavior, adapted to the operational context and meaningful to the operator. The major challenge will therefore be to reconcile inconsistency measurement and human intuition as to its behavior. The formal framework chosen should be rich enough to handle the heterogeneity and imperfection of information, and to reflect different facets of human reasoning. As such, the theory of belief functions will be favored. The approach developed will be tested on real data from AIS (Automatic Identification System), possibly supplemented by information extracted from radar or satellite images, from past events, intelligence reports, patterns-of-life, nautical charts or social networks.
 D. Dubois, W. Liu, J. Ma, H. Prade. The basic principles of uncertain information fusion. An organised review of merging rules in different representation frameworks. Information Fusion 32:12-39, 2016.
 A. Hunter, S. Konieczny. Measuring inconsistency through minimal inconsistent sets. Proc. of the 11th Int. Conf. on Principles of Knowledge Representation and Reasoning (KR’08), pp. 358-366, 2008.
 S. Destercke, T. Burger. Toward an axiomatic definition of conflict between belief functions. IEEE Trans. Syst. Man Cybern. B 43(2):585–596, 2013.
 T. George, N. R. Pal. Quantification of conflict in Dempster-Shafer framework: A new approach. Int. J. of General Systems 24(4):407–423, 1996.
 R. R. Yager. On considerations of credibility of evidence. Int. J. Approx. Reason. 7(1/2):45–72, 1992.
 G. Pallotta, A.-L. Jousselme. Data-driven detection and context-based classification of maritime anomalies. In Proc. of the 18th Int. Conf. on Information Fusion, Washington, D. C., USA, pp. 1152–1159, 2015.
 F. Pichon, A.-L. Jousselme, N. Ben Abdallah. Several shades of conflict. Fuzzy Sets and Systems (2019), https://doi.org/10.1016/j.fss.2019.01.014
 F. Pichon, S. Destercke, T. Burger. A consistency-specificity trade-off to select source behavior in information fusion. IEEE Transactions on Cybernetics 45(4):598-609, 2015.
 N. Ben Abdallah, A.-L. Jousselme, F. Pichon. An ordered family of consistency measures of belief functions. In F. Cuzzolin, S. Destercke, T. Denœux and A. Martin, editors, Belief Functions: Theory and Applications, Proc. of the 5th International Conference, BELIEF 2018, Compiègne, France, September 17-21, 2018, volume 11069 of Lecture Notes in Computer Science, pages 199-207, Springer, 2018.
 N. Ben Abdallah, A.-L. Jousselme, An evidential solution to support reasoning with partially reliable and conflicting sources in maritime surveillance, Under review, 2019.
Supervisor: Frédéric Pichon, Associate Professor (Maître de conférences HDR), Laboratory of Computer Engineering and Automation of Artois (LGI2A)
Co-supervisor: Anne-Laure Jousselme, Scientist, NATO Centre for Maritime Research & Experimentation (CMRE)
Université d’Artois and/or French Région Hauts-de-France
The work will be carried out in the Laboratory of Computer Engineering and Automation of Artois (https://www.lgi2a.univ-artois.fr/spip/?lang=en) located in Béthune, France.
Stays at the NATO Centre for Maritime Research & Experimentation (https://www.cmre.nato.int), located in La Spezia, Italy, may also be necessary.
The candidate must hold a master’s degree (or equivalent) in computer science. Knowledge in artificial intelligence will be an asset.
Send CV, cover letter, grades and rankings of the last two years of studies as well as letters of recommendation to