摘要: |
An accurate knowledge of the pavement surface type is fundamental to efficient pavement management. Knowledge of the surface type is important for tracking the performance of different surfaces and predicting skid numbers and noise. While surface type is already a data element in Pavement Analyst, it lacks the necessary accuracy. Previous Texas Department of Transportation (TxDOT) projects have attempted to populate this field by combining different data sources, but the results have not been implemented because of lack of accuracy. Laser, video, and 3D technologies have made it possible to scan large networks and obtain full coverage on an annual basis. Artificial intelligence and these technologies can be used to predict surface type with an accuracy higher than 90%. This research project will utilize the latest technologies to develop equipment and a methodology for determining surface type. To accomplish this goal, the research team will conduct a literature review to identify potential technologies, evaluate them and determine the technology with the highest potential, develop a set of technical specifications for the system, set a target for the degree of accuracy, develop an experimental design, use the system developed to determine surface type, evaluate its performance, and assemble and deliver a final system. |