IFSTTAR COSYS LEPSIS

Versailles, France

IFSTTAR COSYS LEPSIS

Versailles, France
SEARCH FILTERS
Time filter
Source Type

Fort A.,Ifsttar TS2 LESCOT | Bueno M.,Ifsttar TS2 LESCOT | Fabrigoule C.,University of Bordeaux Segalen | Navarro J.,University Lumiere Lyon 2 | And 2 more authors.
Recherche Transports Securite | Year: 2015

This paper presents a review of the literature to shed light on the impact of distractions on the effectiveness of warning indicating a risk of collision. The rear-end collisions account for a significant share of injury accidents. One of the major contributing factors to these collisions is drivers’ inattention. To reduce the impact of this kind of collisions, driving assistance consisting in alerts have emerged. The effectiveness of these alerts has been evaluated in different situations (simulation, simulator and on road) in non-distracted and distracted drivers. In the latter case, the majority of distraction tasks used involves a diversion of eyes off the road to a device inside the vehicle, coupled or not with manual handling. However, few studies have also examined the impact of distractive tasks having only a cognitive component. The data show that the efficiency of an alert may depend on the interaction between its sensory modality and that of the distractive task. Finally, the effectiveness of these systems can also be affected by the interaction between its reliability and being or not distracted. This review raises research questions that should be further developed to ensure the safety of a larger number of users. © IFSTTAR et Éditions NecPlus 2015.


Dommes A.,IFSTTAR COSYS LEPSIS | Granie M.-A.,IFSTTAR TS2 LMA | Cloutier M.-S.,Center Urbanisation Culture Societe | Coquelet C.,IFSTTAR TS2 LMA | Huguenin-Richard F.,Paris-Sorbonne University
Accident Analysis and Prevention | Year: 2015

To study human factors linked to red light violations, and more generally to safety-related behaviors at signalized crosswalks, the present study combines the collection of observational data with questionnaires answered by 422 French adult pedestrians. Thirteen behavioral indicators were extracted (12 before and while crossing, and red light violation), and the roles of several demographical, contextual and mobility-associated variables were examined. The results of the stepwise logistic regression analyses carried out on each of the 12 behavioral indicators observed before and while crossing revealed that gender had no major impact, but age did, with more cautious behaviors as pedestrians were older. The three contextual variables (group size, parked vehicles, and traffic density), as four mobility-associated variables (driving and walking experiences, self-reported crossing difficulties and falls in the street) were also found to be important factors in safety-related crossing behaviors. A wider logistic regression analysis, made specifically on red light violations with all behavioral indicators observed before and while crossings and the several demographical, contextual and mobility-associated variables put together, showed that red light violations were mostly affected by current situational factors (group size, parked vehicles) and particularly associated with some behavioral patterns (looking toward the traffic, the ground, the light, running and crossing diagonally). The overall results encourage the development of safer pedestrian infrastructures and engineering countermeasures. © 2015 Elsevier Ltd. All rights reserved.


Cavallo V.,IFSTTAR COSYS LEPSIS | Ranchet M.,IFSTTAR COSYS LEPSIS | Pinto M.,IFSTTAR COSYS LEPSIS | Espie S.,IFSTTAR COSYS CODIR | And 2 more authors.
Accident Analysis and Prevention | Year: 2015

The most frequent cause of motorcycle accidents occurs when another vehicle violates the motorcycle's right-of-way at an intersection. In addition to detection errors, misperception of the approaching motorcycle's speed and time-to-arrival is another driver error that accounts for these accidents, although this error has been studied less often. Such misperceptions have been shown to be related to the small size of motorcycles and to their small angular velocity when approaching. In two experiments we tested the impact of different motorcycle headlight configurations in various ambient lighting conditions (daytime, dusk, and nighttime). The participants drove on a driving simulator and had to turn left across a line of vehicles composed of motorcycles and cars. The motorcycles were approaching at different speeds and were equipped with either a "standard" headlight, a "horizontal" configuration (added to the standard headlight were two lights on the rearview mirrors so as to visually increase the horizontal dimension of the motorcycle), a "vertical" configuration (one light on the rider's helmet and two lights on the fork were added to the standard headlight so as to increase the vertical dimension of the motorcycle), or a "combined" configuration (combining the horizontal and vertical configurations). The findings of the first experiment in nighttime conditions indicated that both the vertical and combined configurations significantly increased the gap car drivers accepted with respect to the motorcycle as compared to the standard configuration, and that the accepted gaps did not differ significantly from those accepted for cars. The advantage of the vertical and combined configurations showed up especially when the motorcycle's approach speed was high. The findings of the second experiment in dusk and daytime conditions indicated similar patterns, but the headlight-configuration effect was less pronounced at dusk, and nonsignificant during the day. The results are discussed with regards to possible applications for motorcycles. © 2015 Elsevier Ltd. All rights reserved.


PubMed | IFSTTAR TS2 LMA, Center Urbanisation Culture Societe, IFSTTAR COSYS LEPSIS and Paris-Sorbonne University
Type: | Journal: Accident; analysis and prevention | Year: 2015

To study human factors linked to red light violations, and more generally to safety-related behaviors at signalized crosswalks, the present study combines the collection of observational data with questionnaires answered by 422 French adult pedestrians. Thirteen behavioral indicators were extracted (12 before and while crossing, and red light violation), and the roles of several demographical, contextual and mobility-associated variables were examined. The results of the stepwise logistic regression analyses carried out on each of the 12 behavioral indicators observed before and while crossing revealed that gender had no major impact, but age did, with more cautious behaviors as pedestrians were older. The three contextual variables (group size, parked vehicles, and traffic density), as four mobility-associated variables (driving and walking experiences, self-reported crossing difficulties and falls in the street) were also found to be important factors in safety-related crossing behaviors. A wider logistic regression analysis, made specifically on red light violations with all behavioral indicators observed before and while crossings and the several demographical, contextual and mobility-associated variables put together, showed that red light violations were mostly affected by current situational factors (group size, parked vehicles) and particularly associated with some behavioral patterns (looking toward the traffic, the ground, the light, running and crossing diagonally). The overall results encourage the development of safer pedestrian infrastructures and engineering countermeasures.


PubMed | IFSTTAR COSYS CODIR and IFSTTAR COSYS LEPSIS
Type: | Journal: Accident; analysis and prevention | Year: 2015

The most frequent cause of motorcycle accidents occurs when another vehicle violates the motorcycles right-of-way at an intersection. In addition to detection errors, misperception of the approaching motorcycles speed and time-to-arrival is another driver error that accounts for these accidents, although this error has been studied less often. Such misperceptions have been shown to be related to the small size of motorcycles and to their small angular velocity when approaching. In two experiments we tested the impact of different motorcycle headlight configurations in various ambient lighting conditions (daytime, dusk, and nighttime). The participants drove on a driving simulator and had to turn left across a line of vehicles composed of motorcycles and cars. The motorcycles were approaching at different speeds and were equipped with either a standard headlight, a horizontal configuration (added to the standard headlight were two lights on the rearview mirrors so as to visually increase the horizontal dimension of the motorcycle), a vertical configuration (one light on the riders helmet and two lights on the fork were added to the standard headlight so as to increase the vertical dimension of the motorcycle), or a combined configuration (combining the horizontal and vertical configurations). The findings of the first experiment in nighttime conditions indicated that both the vertical and combined configurations significantly increased the gap car drivers accepted with respect to the motorcycle as compared to the standard configuration, and that the accepted gaps did not differ significantly from those accepted for cars. The advantage of the vertical and combined configurations showed up especially when the motorcycles approach speed was high. The findings of the second experiment in dusk and daytime conditions indicated similar patterns, but the headlight-configuration effect was less pronounced at dusk, and nonsignificant during the day. The results are discussed with regards to possible applications for motorcycles.

Loading IFSTTAR COSYS LEPSIS collaborators
Loading IFSTTAR COSYS LEPSIS collaborators