Driving is a complex activity involving factors associated with the driver, vehicle, and road/traffic environment. This report reviews performance analysis and measurement technology that can be used to further understand the role driver, vehicle, and environmental factors play in accident causation, and ultimately the extent to which vehicles can be designed to maximize accident avoidance capability. Vehicle based accident avoidance countermeasures can allow the driver to detect hazards sooner, thus providing more time for evasive action as well as permitting more effective evasive maneuvering. Review of analytical and experimental procedure for analyzing accident avoidance performance of the driver/vehicle system is the main thrust of this work. Discussion of accident taxonomies and driver behavior are provided as introductory material. Generic hazards that lead to accidents are presented in Section II along with important independent and dependant variables associated with the driver, vehicle, and road/traffic environment. Analytical and experimental research approaches are also summarized including considerations for assessing their applicability to given research problems. Simulator and instrumented vehicle testing are the primary experimental techniques considered here, and key considerations and approaches are reviewed in Section III. Performance measures and procedural considerations are also discussed. Tradeoff considerations in the use of simulator vs. instrumented vehicle techniques are presented in Section IV along with an example research problem involving traffic conflict scenarios. Several appendixes provide more detail on accident taxonomies, conflict models, driver/vehicle analytical models, site visits made during the project, and test course apparatus useful in instrumented vehicle testing. (Author/publisher)
Abstract