❤️ XR
The Georgia Department of Transportation (GDOT) manages over 80,000 miles of road centerlines, which amounts to about one-third of the total #roads in the U.S. state. To optimize their resources, GDOT hired software and design company, Phocaz Inc. to automate the extraction of road features such as lane widths and numbers, shoulder widths and locations, sidewalks, crosswalks, islands, bike lanes, and more. The project involved unlocking asset data from legacy files, integrating new multi-sourced #data, and creating user-based automated workflows. Phocaz generated synthetic images of road signs which it could upload to the #digitaltwin and created an AI agent capable of virtually driving along the lanes in the twin to detect where lane markings were. Applying Bentley-based solutions simplified the CAD-GIS workflow, reducing the difficulties of creating a machine-learning model. Automating and digitizing the process of identifying highway assets and their locations was faster and cheaper than manual workflows, and provided more precise results. This use of #AI and #ML provided additional context to stitch the data together and provide a comprehensive digital representation of Georgia’s road network. #engineering #iTwin #CAD #geospatial