AI Powers First Continuous Traffic-Monitoring System

The Central Nippon Expressway Company (NEXCO Central) is the first operator to utilize an AI-enabled traffic-monitoring system to optimize vehicle patterns in Japan.

NEC Corporation developed an analytical system for NEXCO Central, leveraging AI technology and fiberoptic sensing to predict speed, location, and travel direction. The vibrations from the road vehicles provide continuous data to predict traffic conditions. Sensing devices are attached to the existing optical fiber network used for communication along the expressways.

Existing highway control systems use cameras or point sensors with inadequate peripheral vision to conduct constant monitoring of incidents and traffic congestion. A vast quantity of these sensors is necessary for continuous observation, which can be expensive.

The company’s analytical AI platform performs accurate traffic monitoring by observing vehicle patterns and speeds to predict trajectories. Probe lights, vibration generation, and optical fiber sensing are all factors that help capture a real-time picture of traffic patterns. The algorithm can predict vehicle speeds for every mile of the highway. This enables a clearer prediction of traffic flow and accidents.

The AI model “learned” from synthetic input points. Environmental noises didn’t impact the accuracy.  With multiple iterations of vehicle trajectories, NEC’s system can observe and monitor high-density traffic patterns over a large highway system.

In the past, one of NEC’s successful joint projects with Verizon involved the monitoring of traffic flow and identification of cracks in poles, using optical fiber sensing technology. The endeavor led NEC to embark on transforming optical fiber cables to a sensor system.

These initiatives are part of NEC’s Safer Cities program, focused on leveraging digital technologies for road safety.

This article first appeared in IoT World Today’s sister publication AI Business.