Leveraging AI for Pavement Management
Pavement Field Inspections and Inventory
We mount GoPro Cameras on the hoods of vehicles and collect high-definition road video. Our computer vision models automatically tag the 20 asphalt distresses and 19 concrete distresses specified in the ASTM-D6433 standard in an objective and standardized manner. Our experienced civil engineering team then performs quality control.
Maintenance and Rehabilitation Strategies
Our inspection data can be directly imported into Streetsaver or PAVER/MicroPAVER without any manual effort. From there, our civil engineering team can provide core insights into what kinds of repairs are best to complete for particular road sections and which repairs to prioritize first. We also specialize in preventative maintenance repair strategies.
GIS Visualizations and Budget Analysis
From the Pavement Condition Indices (PCIs) we compute for every road section, we can create shapefiles and KML files to visualize road network conditions on platforms such as Google Earth. Our engineers can also give the city recommendations on budgeting the actual repairs using existing modules on Streetsaver/PAVER. We also present the city with a library of high quality video data for every road section.
Beginning as a research project at Stanford University, Roadata is a computer vision and artificial intelligence company that specializes in assessing road pavements. Leading machine learning researchers developed an end-to-end pipeline that begins with collecting high definition roadway video and using the machine learning models we develop to automatically identify all roadway distresses according to the full ASTM D6433 standard. The data we collect can be directly imported into Streetsaver/PAVER and is highly interpretable and standardized for cities to perform budgeting and repair planning.
We are fast, cost-efficient, standardized, and provide extremely high quality data.
MEET THE TEAM
Former Google Software Engineer
Ex-Researcher at Stanford DAWN Lab
Published Machine Learning Papers in IEEE
Former Machine Learning Engineer at K-Motion
Researcher at Stanford Computer Vision Labs
Published Computer Vision Papers in Elite International Conferences (CVPR)
For any inquiries, questions or commendations, please call: 510-480-2596 or fill out the following form
673 Escondido Rd
Stanford, CA 94305
To apply for a job with Roadata, please send a cover letter together with your C.V. to:
Get a quote: 510-480-2596