This report describes an experimental investigation performed at live intersections to gather infrastructure-based naturalistic driver approach behavior data. This data was collected and analyzed with the goal of understanding how drivers approach intersections under various speeds and environmental conditions. Six stop-controlled intersection approaches across five intersections in the New River Valley, Virginia area were selected for data collection. The sites were selected based on the intersection characteristics and crash statistics. Data was collected from each site for at least two months resulting in over sixteen total months of data. A data acquisition system was devised and implemented to obtain the first intersection data set with fidelity sufficient for developing intersection collision avoidance threat assessment algorithms. An explorative analysis of driver stopping behavior and vehicle trajectories was also performed. Results indicate that an intersection collision system for stop-controlled intersections is feasible. Avenues for future research and potential uses of this new database are highlighted.