<p> The increasing trend in crashes and consequent fatalities due to distracted driving is a growing safety concern in our road network. With rapid advancement in cellphone and in-vehicle technologies along with driver’s inclination to multitasking, the number of crashes due to distracted driving are further on the rise. Some previous studies attempted to detect distracted driving behavior in real-time to mitigate this issue. However, these studies mainly focused on detecting either visual or cognitive distractions, while most of the real-life distracting tasks involve driver’s visual, cognitive, and physical workload, simultaneously. Additionally, previous studies frequently used eye, head, or face tracking data, although current vehicles are not equipped with technologies to acquire such data. To address above issues, this driving simulator study focused on developing algorithms for detecting specific distraction tasks using only vehicle control and driving performance measures. Specifically, algorithms were developed to detect two distracting tasks – texting and eating/drinking. Three data mining techniques were explored – Linear Discriminant Analysis (LDA), Logistic Regression (LR), and Support Vector Machine (SVM). SVM algorithms found to outperform LDA and LR, which detected texting and eating/drinking distraction with an accuracy of 84.33% and 79.53%, respectively. This study may provide useful guidance to successful implementation of distracted driver detection algorithm in Vehicle to Infrastructure (V2I) and Vehicle to Vehicle (V2V) communication, as well as to auto manufacturers interested in integrating distraction detection systems in their vehicles. </p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:10127213 |
| Date | 09 August 2016 |
| Creators | Atiquzzaman, Md |
| Publisher | Southern Illinois University at Edwardsville |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0017 seconds