← All projects

Road AI

Automatically assesses road surface condition from imagery using computer vision, replacing slow manual inspection of defects such as potholes.

Lead researcher
Dympna O’Sullivan
Institution
Technological University Dublin
Road AI

About the project

Road infrastructure maintenance presents major financial and safety challenges globally. In Ireland alone, over 60% of drivers report vehicle damage from potholes, while Transport Infrastructure Ireland allocated €35 million in 2023 to national road maintenance. The manual inspection of road conditions is time-consuming and resource-intensive, particularly in the defect identification and assessment stages. Road AI helped develop a holistic, intelligent software application for automatic road pavement condition assessment. Leveraging computer vision, image processing, machine learning and deep learning techniques

The key objectives of Road AI are:

1.Detecting and evaluating road defects from video data.
2.Classifying road segments using PSCI (Pavement Surface Condition Index) ratings.
3.Providing a user-friendly visual interface for engineers and surveyors.

Road AI established a deployable tool applicable to both Irish and US road contexts and a comprehensive application integrating deep learning object detection, PSCI classification algorithms, and high-definition LCMS/LiDAR data analysis.

Last updated: 9 September 2026

Project team

Dympna O’Sullivan

Lead researcher
Technological University Dublin

Ready to start collaborating?

Let's discuss how The ARC Hub for ICT can help accelerate your digital transformation initiatives and bring your innovative ideas to life.