Identifying and Quantifying Intersection Obstruction and Its Severity Using LiDAR Technology and GIS Spatial Analysis

摘要

Identifying obstructions at intersections is crucial for roadway safety improvement. However, fieldmeasurements used to identify intersection obstructions are labor intensive, costly, and dangerous. Thisstudy explores the feasibility of identifying obstructions at intersections by using light detection andranging (LiDAR) data already available for the flood analysis. Because of its capability of modeling 3Dobjects, emerging LiDAR technology provides a good opportunity for identifying intersectionobstructions. The objective of this paper is to propose a methodology that can identify intersectionobstructions using LiDAR technology, 3D terrain modeling, a plane-of-sight (POS), and GeographicInformation System (GIS) spatial analysis. In addition, this methodology can quantify the severity levelsof intersection obstructions, and it can, potentially, be used for network-level safety improvementprioritization. Using actual LiDAR data that are already available from Savannah, Georgia, experimentaltests are conducted to identify obstructions at intersections. The intersection obstructions, identified usingthe proposed methodology, are compared with those obtained using the previous 100-ft (30.5 m) line-of15sight (LOS) method. Results show the proposed method correctly identifies 92% of intersectionobstructions, compared to 64% identified by the LOS method. The severity levels of intersection sightdistance obstructions can also be quantified. Results show the proposed methodology provides a feasibleplatform and method for transportation agencies to utilize emerging LiDAR technology to identifyintersection obstructions.