Measuring the safety impact of road infrastructure systems on driver behavior: Vehicle instrumentation and real world driving experiment

Featured in this pilot experimental study is the construction and design of an instrumented vehicle that is able to capture vehicle trajectory data with an extremely high level of accuracy and time resolution. Once constructed and properly instrumented, the various data collection systems were integrated with one another and a driving experiment was conducted on northern Virginia roadways with 18 participants taking part in the study. Trajectory data were collected for each of the drivers as they traversed a predefined loop of four roadway segments with varying numbers of lanes and varying shoulder widths. Data collected from the experiment were then used to calibrate the parameters of the prospect theory car-following model through a genetic algorithm calibration procedure. Once all model parameters were successfully calibrated, significance testing was carried out to determine the impacts that the varying roadway infrastructure had on driving behavior. Results indicated that there were significant changes in behavior when comparing one lane roadways to their two lane counterparts—specifically in cases where the roadway featured a wide shoulder. Additional testing was conducted to ensure that there was no variation based on gender, as nine study participants were female and nine were male. The successfulness of this first study conducted with the newly constructed instrumented vehicle creates the opportunity for a variety of additional studies to be conducted in the future.