详情

原文传递 Assessing Fitness to Drive Among Older Drivers:comparative Analysis of on-Road and Simulator Based Driving

论文题名：	Assessing Fitness to Drive Among Older Drivers:comparative Analysis of on-Road and Simulator Based Driving
关键词：	Fitness to drive;ON-ROAD;SIMULATOR;ASSESSING FITNESS
摘要：	Elderly drivers being representative of unique cohort needs safe mobility and drivability.Enabling older drivers to remain mobility while at the same time not risking public safety,adapting new technologies for assessing their fitness to drive is a dire need of the hour. However in field of transportation, ageing has been a challenging for this specific group because of their functional limitations which can discourage the desire of the elderly to drive further.Drivers self-reporting has been widely accepted for evaluating drivers ability and their fitness to drive.But the competencies of drivers can be more threatening and uncertain when they are asked to self-evaluate themselves if they conceal their impairments and lapses deliberately. To meet the requirements of assessing fitness to drive among older adults,recently several studies have proved the validity of simulator studies.This study adheres to the opinion of developing a method to assess fitness to drive using combination of various assessments and driving simulator rides.The goals are a)to evaluate older drivers'fitness to drive based on different assessments b) to make a comparative analysis of three different classifiers on the basis of their predictive performance c) to determine which type of assessment, or combination of assessment provides the best prediction of the fitness to drive d)to determine the validity of driving simulator performance in predicting the fitness to drive.The criterion of fitness to drive was determined in an on-road driving assessment by specialized fitness to drive evaluator from Center for Determination of Fitness to Drive and Car Adaptations(CARA).The validity of predictor to determine fitness to drive was analyzed by using Random Forest, Support Vector Machine,Gradient Boosted Machine and Receiver Operating Curve Analyses.All models could produce test-takers'fitness to drive,but the model combining seven finally-selected significant predictors of the three assessments, yielded the best prediction for fitness to drive.This model reached up to an accuracy of 92.8% on average, making the approach of combining predictors from different assessments acceptable and strongly valid for assessing fitness to drive among older drivers.
作者：	ONAIRA ZAHOOR
专业：	Master of Engineering
导师：	Yongjun Shen
授予学位：	硕士
授予学位单位：	东南大学
学位年度：	2019
正文语种：	中文