Search

Transport Research Arena (TRA) Conference, Lisbonne, PORTUGAL, 14-/11/2022 - 17/11/2022; Investigating psychological characteristics and dynamic behaviors can provide insight in the behavior of at-risk drivers. Achievement goals in driving have recently been studied to assess driver motivation. The four achievement goals in driving are mastery-approach, performance-approach, mastery-avoidance, and performance-avoidance. Naturalistic Driving Studies provide access to objective measurements of driving (i.e., speed, acceleration). Three dynamic criteria have been developed to objectively characterize driver behavior. The aim of this study was to examine the predictive role of achievement goals on objective driving behaviors. During 8 months, 4,626,379 km of 299 drivers was recorded, and simultaneously, the Achievement Goals in Driving Questionnaire was completed. Mastery-approach goals seem to be the most protective goals in driving, as opposed to performance-approach goals.

International audience; Investigating psychological characteristics through self-reported measures (e.g., anger, sensation seeking) and dynamic behaviors through objective measures (e.g., speed, 2D acceleration, GPS position etc.) may allow us to better understand the behavior of at-risk drivers. To assess drivers' motivation, the theoretical framework of achievement goals has been studied recently. These achievement goals can influence the decision-making and behaviors of individuals engaged in driving. The four achievement goals in driving are: seeking to improve or to drive as well as possible (mastery-approach), to outperform other drivers (performance-approach), to avoid driving badly (mastery-avoidance), and to avoid being the worst driver (performance-avoidance). Naturalistic Driving Studies (NDS) provide access to the objective measurements of data not accessible through self-reported measurements (i.e., speed, accelerations, GPS position). Three dynamic criteria have been developed to characterize the behavior of motorists objectively: driving events, time spent above acceleration thresholds (longitudinal and transversal), and the extent of dynamic demands. All these criteria have been measured in different road contexts (e.g., plain). The aim of this study was to examine the predictive role of the four achievement goals on these objective driving behaviors. 266 drivers (96 women, 117 men) took part in the study, and 4 242 482 km was recorded during 8 months. Simultaneously, they completed the Achievement Goals in Driving Questionnaire. The main results highlighted that mastery-approach goals negatively predicted hard braking and the extent of dynamic demands on plain and hilly roads. Mastery-approach goals seem to be the most protective goals in driving. Future research on the promotion of mastery-approach goals in driving may be able to modify the behavior of at-risk drivers.

Livrable L5 présentant les travaux du workpackage 5 du projet de recherche DYMOA+. L'objectif de la première partie de ce Work Package n°5 (WP5) est d'améliorer les critères de détection d'un incident pour un 2RM définis et utilisés dans le projet antérieur DYMOA. Les choix des variables et des seuils retenus pour ces critères semblent bien adaptés pour recueillir des évènements utiles pour le diagnostic d'infrastructure. En revanche ces critères s'avèrent moins efficaces pour différencier ces évènements d'incidents et surtout pour détecter les incidents "réels". Beaucoup trop de faux positifs sont enregistrés et de plus, des faux négatifs nous échappent certainement. Le but de ces travaux est donc d'améliorer ces critères ou d'en identifier de nouveaux, permettant si possible, de détecter des incidents "réels" et de les discerner des "fortes sollicitations". Pour cela il est probablement pertinent de combiner les variables et/ou de rajouter des notions de durée de dépassement de seuil. Une attention particulière est portée sur les variations de sollicitations, signes d'un effet de surprise, et annonciateurs de la présence d'une situation non souhaitée, plus proche de la définition d'un incident. Nous avons employé des méthodes utilisant des algorithmes d'apprentissage supervisé appliqués sur des jeux de données du projet DYMOA et d'autres données complémentaires acquises dans le cadre de ce projet DYMOA+. Ces méthodes donnent des résultats prometteurs qui restent à approfondir. La deuxième partie de ce WP5 présente l'exploitation complémentaire de données recueillies à la toute fin du projet DYMOA. Les données de conduite naturaliste de l'expérimentation DYMOA, acquises pendant une période de 18 mois avec 26 motos privées dans trois régions de France, ont été exploitées pour évaluer les niveaux de sollicitations réelles des 2RM. Il s'agissait plus particulièrement de données agrégées fournies pour l'ensemble des trajets, appelées « synthèses de parcours » et croisant les données accélérométriques et gyrométriques ainsi que les niveaux de vitesses. Cette étude vient compléter les résultats du livrable 3.3 de DYMOA, qui n'exploitait qu'une partie des données, concernant environ 3200 trajets. Sur 7500 trajets collectés, environ 6500 trajets ont pu être exploités pleinement, en correspondance avec les traces GPS qui permettent notamment de calculer les distances parcourues. La distance globale parcourue lors de ces trajets est d'environ 88000 km. Les résultats obtenus sur le jeu complet des données concernent en premier lieu les distributions globales, en temps passé, des accélérations longitudinale, latérale et verticale, et des vitesses de rotation tangage, roulis et lacet et de la vitesse, puis les valeurs extrêmes atteintes au moins une fois par les motocyclistes pour ces mêmes paramètres. La variabilité des comportements est ensuite illustrée allant d'une conduite souple à une conduite sportive. Les motocyclistes subissent rarement des niveaux élevés de sollicitations dynamiques. En accélération et décélération notamment, ils ne dépassent +-4 m/s² que 0,5% du temps. En ce qui concerne les virages, leur vitesse de roulis ne dépasse 20°/s que 0,8% du temps et leur vitesse de lacet que 2,1% du temps. Si l'on exclut le temps passé à l'arrêt ou à très faible vitesse (moins de 5 km/h), les motocyclistes passent 80 % de leur temps en dessous de 90 km/h et 3,2 % du temps au-dessus de 130 km/h, avec seulement 0,3 % du temps au-dessus de 150 km/h. Le comportement est très variable d'un motocycliste à l'autre, même sur les mêmes itinéraires. Ces données permettent de caractériser la dynamique des motos et de discriminer les différents comportements des motards.

The abnormally high mortality rate of Powered Two-Wheelers (PTW) is probably related to their vulnerability, but also to their behavior and interactions with other road users. There is a need to collect exposure data on PTWs to better understand uses and determinants involved in the mechanisms of accidents since there are almost no naturalistic studies on motorcyclists as on motorists. This paper aims to characterize the actual use dynamic capabilities of PTWs by motorcyclists through objective measures. In 2016-2018, an experiment was carried out to acquire Naturalistic Riding Data from 26 private motorcycles implemented with an Event Data Recorder. For 18 months, in three regions of France, the devices collected aggregated data on each ride thanks to accelerometers and gyrometers, and continuous speeds and trajectories, thanks to GPS. This paper presents the global distributions of accelerations, rotation rates and speed, and the extreme values reached at least once. It also illustrates the variability of behaviors: smooth driving versus sporty driving. 6500 travels were exploited, representing 88000 km. The motorcyclists endure rarely high levels of dynamic demands; especially in acceleration and deceleration, they exceed +-4 m/s² only 0.5% of the time. As far as cornering is concerned, their roll rate exceeds 20°/s only 0.8% and their yaw rate 2.1% of the time. Excluding time spent at standstill or at very low speed (less than 5 km/h), motorcyclists spend 80% of their time below 90 km/h and 3.2% of the time above 130 km/h, with 0.3% of the time above 150 km/h. The behavior varies greatly from one motorcyclist to another, even on the same itineraries. These data characterize the dynamics of motorcycles and discriminate different behaviors of riders.

During naturalistic driving studies on Powered Two-Wheelers (PTW), a lot of data is acquired and difficulties appear to analyze all of it. So it seems interesting to focus on incident detection to shorten the driving sequences to study. Furthermore, a lot of studies were conducted concerning accidents, hence the decision to focus on near-crashes. This work deals with the determination of the most adequate PTW's dynamics parameters and their thresholds level allowing incidents detection. More than 60 experimental tests were performed on track (braking, slaloms, avoidance manoeuvres...) and on road with an instrumented motorcycle. Data recorded were speed, accelerations, rotation speeds, steering angle, steering torque, lean angle, brake pressure and trajectory. The method consists to observe these parameters in parallel with the feeling of the rider when performing manoeuvres. For example, the rider was not considering engaging strongly the motorcycle capacities for the slalom at 30km/h, while the avoidance manoeuvre at the same speed was considered as an emergency manoeuver. Then, criteria and thresholds have been determined to highlight the suddenness of a manoeuvre and the urgency of a situation. For the longitudinal dynamics, the retained conditions are: Longitudinal Speed greater than 10 km/h and Longitudinal Acceleration, under braking only, less than -6m/s². Concerning lateral dynamics, two possible criteria were retained: norm of Rotation Speed vector (Roll+Yaw+Pitch) greater than 80°/s or norm of Rotation Acceleration vector (Roll+Yaw+Pitch) greater than 300°/s². These criteria were afterwards validated on road travels were the rider ensures not to be confronted to any emergency situation. This study allows determining the PTW dynamics parameters that make possible the incident or near-crashes detection. About 30 vehicles were then instrumented to collect a large amount of data on the use of dynamics capacities. In the future these data collection will permit to refine these thresholds values.