Steering or braking avoidance response in SHRP2 rear-end crashes and near-crashes: A decision tree approach
نویسندگان
چکیده
The paper presents a systematic analysis of drivers’ crash avoidance response during crashes and near-crashes developed machine learning-based predictive model that can determine driver maneuver using pre-incident behavior driving context. We analyzed 286 naturalistic rear-end from the SHRP2 study. All events were manually reduced face video (face forward) kinematic responses. In this paper, we new reduction variables enhanced understanding gaze roadway attention these events. These features reflected how event criticality, measured time to collision, related (secondary behavior, behavior), perception (physical reaction maneuver). imperative such relations was validated random forest- (RF) based classifier, which efficiently predicted if going brake or change lane as an maneuver. RF presented in effectively explored nonlinear patterns data highly accurate (∼96 %) its prediction. A further showed six played pivotal role decision logic. included last glance duration before event, eccentricity, ‘eyes on road’ immediately instance criticality when perceives threat well acknowledge threat, possibility escape path adjacent lane. Using partial dependency plots, also different thresholds feature determined intention. context, unified structure predict their response. To best our knowledge, is first effort where large-scale (crashes near crashes) for prediction key behavioral contextual factors contribute
منابع مشابه
Analyses of rear-end crashes based on classification tree models.
OBJECTIVE Signalized intersections are accident-prone areas especially for rear-end crashes due to the fact that the diversity of the braking behaviors of drivers increases during the signal change. The objective of this article is to improve knowledge of the relationship between rear-end crashes occurring at signalized intersections and a series of potential traffic risk factors classified by ...
متن کاملEvaluating the Relationship Between Near-Crashes and Crashes: Can Near-Crashes Serve as a Surrogate Safety Metric for Crashes?
Traffic Safety Administration, in the interest of information exchange. The opinions, findings, and conclusions expressed in this publication are those of the authors and not necessarily those of the Department of Transportation or the National Highway Traffic Safety Administration. The United States Government assumes no liability for its contents or use thereof. If trade names, manufacturers'...
متن کاملMinor Rear Aligned Crashes in the United States: a Pilot Study of 98 Crashes
In the United States there are few available data regarding real world minor rear aligned crashes with struck vehicle delta-V less than 15 km/h. This study analyzed 98 such crashes in the US involving 258 occupants. Mean struck vehicle delta-V and acceleration were 6.4 km/h and 1.4 g, respectively. Within five weeks post-crash, 105 struck vehicle occupants had 695 ICD-9-CM diagnoses and 399 AIS...
متن کاملRisk factors for 'whiplash' in drivers: a cohort study of rear-end traffic crashes.
'Whiplash' injuries are a common cause of road traffic accident (RTA) related morbidity. However, few epidemiological studies have investigated risk factors associated with the occurrence of such injuries. The purpose of this study was to examine relationships between vehicle and occupant factors and the incidence of 'whiplash' in drivers with recent experience of a rear-end impact in an RTA. A...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Accident Analysis & Prevention
سال: 2021
ISSN: ['1879-2057', '0001-4575']
DOI: https://doi.org/10.1016/j.aap.2021.106055