Drivers who have had to use the same route for years have memorized which points are in poor condition or pose a danger. However, it usually takes months for these problems to be reported to the authorities and included in the repair plan. A new Honda AI-focused pilot project, aiming to eliminate this delay, is currently in the testing phase, particularly in Ohio. This system automatically detects problems on the road using data collected by vehicles and instantly reports them to the relevant authorities.
Honda's collaboration with DriveOhio enabled the direct integration of data obtained through sensors and cameras into road maintenance processes. The tested vehicles were used on approximately 3 miles of routes in both urban and rural areas, and driving was conducted in varying weather conditions and time zones. This allowed for detailed observation of the system's real-world performance.
Details that Identify Road Problems
During the tests, the vehicles used advanced camera and LiDAR sensors to detect various problems on the road. The system focused particularly on the following areas:
- Potholes and damaged asphalt surfaces
- Missing or damaged railings
- Traffic signs that are obstructing visibility or are tilted.
This list clearly outlines the system's coverage and shows the scope of data collected compared to manual checks.
Accuracy Level of Artificial Intelligence
Honda AIThe system stands out with its high accuracy rates in the results obtained in the project. The system correctly identified traffic sign problems with 99% accuracy and guardrail problems with 93% accuracy. Potholes and road surface defects were identified with nearly 90% accuracy. These rates represent a significant improvement compared to traditional methods and reduce the workload of maintenance teams by keeping the number of false alarms low.
ZamReal-time Reporting and Process Acceleration
The collected data is directly integrated into the Ohio Department of Transportation's systems, allowing officials to quickly access details of problems. This enables them to pinpoint the exact location and nature of the problem without needing to send teams to the field. This approach offers advantages in terms of both time and cost. According to Honda, this system produces more consistent results compared to human-based reporting methods.
Although the system is currently used in limited areas, Honda plans to scale up the data in an anonymized form. As information from vehicles becomes more widespread, proactive and early problem detection in road maintenance will become possible. According to the company's estimates, this innovation could save over $4,5 million in Ohio within a few years. One of the most important advantages is that drivers can contribute to solving infrastructure problems during their own journeys without having to do anything extra.
