Keywords: Signal processing, AI, radar, video, data fusion

Summary

The increasing number of vehicles on the road today implies a growing number of road accidents. It has been shown that the physical and physiological state of the car driver is a significant cause of accidents [1]. It has also been demonstrated that cognitive fatigue and distraction are also responsible for many road accidents [2].

This thesis focuses on the analysis and automatic recognition of the driver's state in a context aimed at preventing risky situations. The study proposes the development of a system that processes physiological data from non-contact sensors (radar and camera). The challenge of the thesis lies in ensuring the reliability and robustness of the proposed system in dealing with disturbances caused by the vehicle's environment and the driver's movements. In this context, challenges persist concerning the reliability of video data processing. Additionally, the utilization of radar data remains a significant difficulty in a real-world environment.

Objectives

Initially, to achieve these objectives - which excludes any operational use of contact sensors - an analysis will focus on the relevance of various metrics representing physical and cognitive states to be considered in such a context. Subsequently, we aim to develop, using machine learning techniques, an advanced multimodal analysis model, effectively merging features derived from physiological and visual signals. It is also noteworthy that, at the end of the thesis, we plan to share with the scientific community the database consisting of synchronized radar and vision sensor signals associated with driving scenarios.

Additional information

The successful candidate will be a member of the CERI Systèmes Numériques (Center for Digital systems) of IMT Nord Europe and COSYS – LEOST lab of Universite Gustave Eiffel. The two entities (IMT NE and UGE) are located in Villeneuve d’Ascq, at approximately 1h30 by TGV train from center Paris.

The successful candidate can start as early as October 2024.

Applications are sought from France, EU and international candidates with an outstanding academic background. The applicants should hold a Master or Engineer degree (BAC+5) and have a strong background in computer science or signal processing. A good English level in writing, reading and speaking is also required. Finally, having a strong mathematical background and machine/deep learning skills (Python, Keras, etc.) is a definite plus.

How to apply

Interested candidates have to send their detailed CV, academic records (from Bsc to Msc level), at least two academic referees and a short motivation letter via email to the contacts below.

Application deadline: May 15th, 2024

Applications will be received until May 15th, 2024 and will be considered as they arrive, therefore early application is highly encouraged.

Contacts:

• ·IMT Nord Europe

Halim Benhabiles - halim.benhabiles@imt-nord-europe.fr

Jose Mennesson - jose.mennesson@imt-nord-europe.fr

• ·Université Gustave Eiffel

Fouzia Boukour - fouzia.boukour@ifsttar.fr

David Sodoyer - david.sodoyer@ifsttar.fr

References

[1] Magaña, V.C.and all, « The Effects of the Driver’s Mental State and Passenger Compartment Conditions on Driving Performance and Driving Stress ». Sensors 2020, 20, 5274.

[2] Mick Salomone, « Analyse comportementale et électrophysiologique de l'impact de la fatigue cognitive sur les capacités d'adaptation », 11 mars 2021,Université d’Aix Marseille.