Vehicle-mounted AI edge cameras and computer vision combined with centimeter-level RTK positioning enable real-time detection of road hazards and urban infrastructure conditions while vehicles are on the move

A.I.MATICS, a specialist in AI-powered computer vision led by CEO Hoon Lee, is partnering with LG Uplus to launch an AI-driven urban data initiative.

A.I.MATICS announced on the 22nd that it will provide its proprietary vehicle-mounted AI edge cameras and computer vision models to build a system capable of detecting and analyzing road hazards and urban data in real time using vehicles that travel through cities every day, including buses, taxis, and delivery vehicles.

In the project, A.I.MATICS’ technology will serve as the “eyes and brain” for urban safety. The company has developed both the AI edge cameras installed on vehicles and the computer vision models that identify potholes and infrastructure assets in captured footage. Its cameras use on-device AI processing, analyzing video directly on the device without relying on external servers and enabling road hazards to be identified in real time as vehicles encounter them.

The precise location of each detected hazard is determined using LG Uplus’ Real-Time Kinematic (RTK) positioning technology, which reduces positioning errors to the centimeter level. As soon as a vehicle detects road surface damage, the system records its precise location, allowing the information to be linked directly to municipal maintenance operations. Unlike conventional GPS, which can have positioning errors of up to 15 meters, RTK minimizes error by connecting ground-based reference stations with dedicated terminals in real time.

The combination of these technologies has direct implications for everyday urban safety. As vehicles travel through the city, they can automatically identify and report hazards such as potholes, which can damage tires and cause accidents, as well as clogged storm drains that can contribute to flooding during heavy rain, malfunctioning streetlights, and damaged manhole covers.

The system goes beyond detecting safety hazards. It can also register and monitor a wide range of urban infrastructure, including street signs, traffic facilities, unauthorized banners, and fire safety equipment. By detecting subtle changes in these assets over time, the system can help improve the efficiency of infrastructure operations and maintenance.

The data collected can ultimately support broader smart city solutions for integrated public infrastructure management. Potential applications extend beyond municipal road maintenance departments to public agencies responsible for inspecting and managing infrastructure assets.

The approach also addresses blind spots in conventional urban monitoring. Fixed CCTV cameras can cover only limited areas, while traditional road inspections often depend on personnel conducting visual checks and manually recording conditions. By turning vehicles already traveling throughout the city into mobile data collection platforms, municipalities can monitor a much wider area at greater frequency and proactively identify hazards—even on smaller streets—before they lead to public complaints or accidents.

Officials can review the latest condition of roads and infrastructure before conducting on-site inspections and track changes in individual assets over time, enabling more efficient and data-driven maintenance decisions.

The initiative also introduces a scalable recurring-revenue business model. Rather than relying solely on one-time hardware sales, A.I.MATICS will generate recurring software license revenue following the deployment of its AI edge cameras. The company expects approximately KRW 20 billion in revenue from the collaboration, with the recurring software revenue base expanding as more vehicles adopt the system.

The market outlook is also strong. The global AI computer vision market is projected to reach approximately USD 63.5 billion by 2030. Within this market, the vehicle vision-based road and pothole detection segment is expected to grow from USD 3.8 billion in 2025 to USD 11.2 billion by 2034, representing a compound annual growth rate of 12.8% and highlighting its growing importance as a core component of smart city infrastructure.

“The market for continuously observing cities and digitally recording their conditions is only beginning to emerge, and the value of this data will increase as it accumulates,” said Hoon Lee, CEO of A.I.MATICS. “With our proprietary computer vision technology, we aim to establish an early leadership position in this market and help set a new standard for smarter urban operations.”

Building on this collaboration, A.I.MATICS and LG Uplus plan to expand customized, real-time urban data services for local governments nationwide. The companies also plan to leverage accumulated data to develop higher-value applications, including predictive maintenance and digital twin simulations.

Source: AI Times Korea