Explore the vast range of titles available.

A significant number of road accidents are caused by impaired driver attention, such as inattention and more specifically mind wandering. Accordingly, to enhance drivers’ safety, one solution proposed to avoid accidents due to mind wandering is to automate vehicle driving. However, vehicles on the market are not yet fully autonomous, and the driver retains supervisory status and must be able to regain control of the vehicle when asked to do so. This passive supervision can lead to under-activation of the attentional system and mind wandering, hindering the driver's ability to regain control of the vehicle if requested. To address this societal issue, the physiological indicators of mind wandering were investigated in delegated driving. Forty-three participants took part in this experiment. They were seated in a car simulator, and supervised a simulated autonomous driving environment for 20 min. Every 60 to 120 s, they were asked to declare whether their attention was focused on the supervisory task or not. Thus, it permits distinguishing between two attentional situations: being focused on the task or wandering thoughts. There were 11 questionnaires and their order reflected the temporal progression throughout the experiment. Before each declaration, cardiac and electrodermal activities were measured. The Linear Mixed Model analysis showed that the attentional situation was a predictor of the number of phasic skin conductance responses and time was a predictor of the tonic skin conductance level. Individuals had fewer phasic responses when they were in mind wandering state than when they were focused on the task and the tonic level varied over time. The attentional situation, the time, and their interaction were predictors of SDNN reactivity (Standard Deviation of Normal-to-Normal intervals). SDNN varied over time and reactivity was lower in mind wandering. The results from the other cardiac indicators were not significant. These results showed specific cardiac and electrodermal signatures of mind wandering. They are discussed as a consequence of low engagement in the task when drivers are mind wandering. Considering these indicators could enable the detection of mind wandering in autonomous driving before a takeover request.

Research works on operator monitoring underline the benefit of taking into consideration several signal modalities to improve accuracy for an objective mental state diagnosis. Heart rate (HR) is one of the most utilized systemic measures to assess cognitive workload (CW), whereas, respiration parameters are hardly utilized. This study aims at verifying the contribution of analyzing respiratory signals to extract features to evaluate driver's activity and CW variations in driving. Eighteen subjects participated in the study. The participants carried out two different cognitive tasks requiring different CW demands, a single task as well as a competing cognitive task realized while driving in a simulator. Our results confirm that both HR and breathing rate (BR) increase in driving and are sensitive to CW. However, HR and BR are differently modulated by the CW variations in driving. Specifically, HR is affected by both driving activity and CW, whereas, BR is suitable to evidence a variation of CW only when driving is not required. On the other hand, spectral features characterizing respiratory signal could be also used similarly to HR variability indices to detect high CW episodes. These results hint the use of respiration as an alternative to HR to monitor the driver mental state in autonomic vehicles in order to predict the available cognitive resources if the user has to take over the vehicle.

Research works on operator monitoring underline the benefit of taking into consideration several signal modalities to improve accuracy for an objective mental state diagnosis. Heart rate (HR) is one of the most utilized systemic measures to assess cognitive workload (CW), whereas, respiration parameters are hardly utilized. This study aims at verifying the contribution of analyzing respiratory signals to extract features to evaluate driver's activity and CW variations in driving. Eighteen subjects participated in the study. The participants carried out two different cognitive tasks requiring different CW demands, a single task as well as a competing cognitive task realized while driving in a simulator. Our results confirm that both HR and breathing rate (BR) increase in driving and are sensitive to CW. However, HR and BR are differently modulated by the CW variations in driving. Specifically, HR is affected by both driving activity and CW, whereas, BR is suitable to evidence a variation of CW only when driving is not required. On the other hand, spectral features characterizing respiratory signal could be also used similarly to HR variability indices to detect high CW episodes. These results hint the use of respiration as an alternative to HR to monitor the driver mental state in autonomic vehicles in order to predict the available cognitive resources if the user has to take over the vehicle.

The SURCA project (Road user safety and automated driving) has two main objectives. The first is to identify scenarios for interactions between autonomous vehicles and other road users (non-autonomous vehicles, motorised two-whcelers, pedcstrians and cyclists. The second objective is to study how the posture of the occupants (driver and passengers) of a vehicle in autonomous mode affects injury risk. This paper focuses on a description of the project (which is still ongoing) and on its first results, mainly concerning the first objective. The first findings allow us, first of ail, to identify the most relevant accident scenarios (cspecially in terms of accidents) for the introduction of automatcd driving. We have performed a preliminary quantification of the effects of the graduai deployment of autonomous vehicles in traffic on the occurrence of persona! injury accidents. Finally, we will present some initial results on the modelling of how the posture of occupants in vchicles operating in autonomous mode affects their injury risk, and on the analysis of the nccds of eldcrly uscrs and their acceptance of automated vehiclcs.

Vulnerable user safety is an important public-health issue; most of the accidents involving trucks and vulnerable users in urban areas have devastating consequences for users. The goal of the VIVRE2 project (anr05-pdit) is the study and design of a driving-assistance system for truck drivers to significantly reduce the number of these accidents. The design of this system was inspired by the User-Centered Design Approach, based on iterative steps described by Maguire (2001). To plan the human-centered process, a methodology was defined to allow several specification versions. This methodology included tool creations to apply this approach in an industrial context. The first one, named Integration Platform, is dedicated to the integration of the driving assistance inside the driving simulator environment to facilitate the iterative process. The second one, named Dynamic Scenario Generator, is dedicated to the design of the assistance system itself. It allowed the verification that a single system, including active controls, is capable of answering the set of use case requirements. To identify critical-use cases, the context of use was analyzed not only from the point of view of drivers but also, from that of vulnerable users. Requirements were defined to improve the way in which the system is used during the driving task according to task constraints (temporal aspect, complexity, attentional demand). In the design solution, an adaptive technology, taking driver’s behavior into account was used to adapt the HMI alerts and active controls. Multiple evaluations were performed to iteratively improve the functioning of the system. This experience has shown that around the truck driving simulator of Renault trucks, designers, psychologists, and ergonomists were able to work together to implement strategies of driver assistance, to assess the relevance and effectiveness of a homogeneous system for the use cases. The designed system was evaluated using the dynamic driving simulator; its acceptability by the driver, and its effectiveness in critical situations (89 % of the accidents were avoided on all the experiments) were demonstrated.

RésuméLa sécurité des usagers vulnérables une question importante en termes de santé publique : la plupart des accidents impliquant des camions et des usagers vulnérables dans des zones urbaines ont des conséquences dramatiques pour ces derniers. Le projet VIVRE2 ( anr 05- pdit ) avait pour objectif l’étude et la conception de solutions technologiques pour réduire significativement le nombre de ces accidents.Pour concevoir ce système, l’approche User Centred Design , basée sur les étapes itératives décrites par Maguire (2001), a été adoptée en incluant des créations d’outils pour pouvoir l’appliquer dans le contexte industriel. Une plate-forme d’intégration a été conçue pour réduire le temps d’intégration du système d’assistance dans l’environnement de simulateur de conduite et ainsi faciliter le processus itératif. De plus, un générateur de scénario dynamique a permis la conception en laboratoire d’un système d’assistance actif en prenant en compte l’ensemble des cas d’utilisation.Le contexte d’utilisation a été analysé selon le point de vue de conducteurs et selon celui des usagers vulnérables. Les spécifications ont été définies pour améliorer l’utilisation de système dans la tâche de conduite selon les contraintes de cette dernière (l’aspect temporel, la complexité, la demande attentionnelle). Le système, basé sur une technologie adaptative, a été développé en prenant en compte le comportement du conducteur pour adapter les alertes de l’Interface Humain-Machine et les commandes actives.Le système retenu a été évalué sur simulateur dynamique de conduite et a prouvé son acceptabilité par les conducteurs et son efficacité sur des situations critiques sélectionnées (il a permis d’éviter 89 % des situations potentiellement critiques sur simulateur). Development of a truck-driving assistance system to detect vulnerable road users in urban areasVulnerable user safety is an important public-health issue: most of the accidents involving trucks and vulnerable users in urban areas have devastating consequences for users. The goal of the VIVRE2 project ( anr 05- pdit ) is the study and design of a driving-assistance system for truck drivers to significantly reduce the number of these accidents.The design of this system was inspired by the User-Centred Design Approach, based on iterative steps described by Maguire (2001). To plan the human-centred process, a methodology was defined to allow several specification versions. This methodology included tool creations to apply this approach in an industrial context. The first one, named Integration Platform, is dedicated to the integration of the driving assistance inside the driving simulator environment to facilitate the iterative process. The second one, named Dynamic Scenario Generator, is dedicated to the design of the assistance system itself. It allowed to verify that a single system, including active controls, is capable of answering the set of use case requirements.To identify critical-use cases, the context of use was analysed not only from the point of view of drivers but also, from that of vulnerable users. Requirements were defined to improve the way in which the system is used during the driving task according to task constraints (temporal aspect, complexity, attentional demand). In the design solution, an adaptive technology, taken into account driver’s behaviour was used to adapt the hmi alerts and active controls. Multiple evaluations were performed to iteratively improve the functioning of the system.This experience has shown that around the truck driving simulator of RENAULT TRUCKS, designers, psychologists and ergonomists were able to work together to implement strategies of driver assistance, to assess the relevance and effectiveness of a homogeneous system for the use cases The designed system was evaluated using the dynamic driving simulator; its acceptability by the driver, and its effectiveness in critical situations (89 % of the accidents were avoided on all the experiments) were demonstrated.