Driver Impairment Detection and Safety Enhancement Through Comprehensive Volatility Analysis

US Transportation Collection2020PDFTech ReportKhattak, Asad J.Arvin, RaminChakraborty, SubhadeepMelton, ChadClamann, MichaelUniversity of Tennessee, Knoxville. Center for Transportation ResearchUniversity of North Carolina at Chapel HillUnited States. Department of Transportation. University Transportation Centers (UTC) ProgramUnited States. Department of Transportation. Office of the Assistant Secretary for Research and TechnologyUnited States. Department of Transportation. Federal Highway AdministrationCollaborative Sciences Center for Road SafetyUnited StatesAlgorithmsAutomatic data collection systemsBehaviorBiometricsData analysisDriversDriving simulatorsImpaired driversSensorsVirtual realityVolatilityTransportation safetyArtificial intelligenceImpairmentDistracted DrivingCSCRS-R23Collaborative Sciences Center for Road Safety (Grant: 69A3551747113)Final Report: May 2019-Aug 2020The ubiquity of sensors and increasing computational resources has enabled monitoring driver, vehicle, and roadway surroundings. A key objective of this research is to extract useful information from multi-dimensional data streams coming in from sensors. The has developed a framework for obtaining, processing, and analyzing high-frequency multi-dimensional large-scale data using vehicle-based sensors that monitor the driver, vehicle, and roadways. The framework harnesses the data and quantify variations in driver biometrics and behavior, vehicle kinematics, and roadway/environmental conditions utilizing the concept of volatility. The naturalistic driving study data from Strategic Highway Research Program (SHRP-2) are utilized for in-depth analysis on impairment and distracted driving. The associated risks with engagement in such behaviors in terms of occurrence of safety critical events are quantified and discussed. A real-time artificial intelligence framework is developed to harness multi-dimensional data and quantify instantaneous crash risk by monitoring driver biometrics (in terms of distraction), vehicular movements, and instability in driving. Furthermore, steps were taken to review the literature on driver monitoring, as well as conducting driving experimentation in simulated and naturalistic settings, which will contribute to future research.113