| Summary: | The SHRP 2 NDS project was the largest naturalistic driving study ever conducted. The data obtained from the study was released to the researching public in 2014 through the projects InSight webpage. The objectives of this research were to first explore the massive dataset and determine if it was possible to develop prediction models based on several performance measures that could be used to study driver distraction. Time series data on driver GPS speed, lateral and longitudinal acceleration, throttle position and yaw rate were discovered as five appropriate performance measures available from the NDS that could be used for the purpose of this research. Using this data the objective was to predict whether a driver was engaged in any of three specific groups of distracting tasks or no secondary task at all. The specific distracting tasks that were examined included: talking or listening on a hand-held phone, texting or dialing on a hand-held phone, and driver interacting with an adjacent passenger. Multiple logistic regression was the statistical method used to determine the odds of a driver being engaged in one of the secondary tasks given their corresponding driving performance data. The results indicated there were differences in the driving performance measures when the driver was engaged in a secondary task. The intent of this research was to determine if those differences present could be used to develop models that could adequate predict when a driver was engaged in the three secondary tasks of interest. The results of the MLR tests indicate this data could not be used to develop prediction models with statistically significant predictive power. Future work should focus on comparing this researchs results to prediction models developed using an alternative to the multiple logistical regression method.
|
|---|
