Traffic Technology International
Advances in ITS are enabling multi-technology detection systems that create proactive safety networks on the roadways
Deep learning algorithms classify each road user, including vulnerable road users
Every year, more than one million people lose their lives on roadways worldwide, with traffic crashes remaining the leading cause of death for people aged 5-29, according to the World Health Organization. This stark reality makes Vision Zero a global imperative. To achieve it, transportation agencies face a persistent challenge: creating safer roadways requires accurate detection of all road users, in every environment, and under all conditions.
Fortunately, advances in ITS are creating new possibilities. Sophisticated multi-technology detection systems and modern traffic AI, enhanced by deep learning, are delivering detection, data, and automation capabilities that were previously unprecedented.
Sensor fusion at intersections
At intersections, the complexity of mixed traffic demands solutions that can reliably detect both vehicles and VRUs. Inductive loops are no longer enough. Sensys Networks’ MultiSens Intersection combines video AI with wireless in-road sensors to address limitations inherent in single technology approaches.
Video AI provides full-scene awareness, accurately detecting vehicles, motorcycles, bicycles, and scooters across all approach lanes simultaneously, as well as pedestrians in crosswalks. Deep learning algorithms classify each road user and automatically filter adverse weather and lighting conditions, maintaining performance in rain, snow, fog, and glare.
Wireless in-road sensors extend detection to advance or midblock locations where line-of-sight is often limited by trees, roadway curves, or other vehicles. These geomagnetic sensors are immune to environmental factors and provide accurate, lane-specific vehicle detection regardless of weather conditions. In addition, the precision of wireless sensors supports red-light and speed compliance applications.
This fusion of technologies reduces false positives and missed detections compared to single-sensor systems. The result is safer signal timing for VRUs and the collection of reliable road user metrics to support Vision Zero.
Coordinated highway safety
This multi-technology approach also creates a proactive safety network on highways. RTMS Echo radar efficiently monitors up to 12 lanes of traffic to provide real-time speed and volume metrics for variable speed limit systems on managed motorways. Safe speeds are determined by current conditions and displayed on dynamic message signs above the roadway. Wireless sensors installed on the on-ramps pair with this mainline data to enable coordinated ramp metering, smoothing traffic entry. These same sensors on off-ramps trigger queue flushing, preventing dangerous backups.
In Melbourne, coordinated ramp metering produced a 31% decrease in crashes during peak times. In the UK, dynamic speed limits reduced injury accidents by 18%.
31% The reduction in peak time crashes in Melbourne resulting from coordinated ramp metering
AI-powered incident detection
When incidents occur in high-risk environments like highways, bridges, and tunnels, response time is critical. Citilog’s Automatic Incident Detection (AID) system detects crashes, stopped vehicles, debris, and pedestrians within seconds. Pre-trained on a vast dataset of real-world incidents, the deep learning algorithms filter environmental factors, reducing false positives by a factor of 10, allowing operators to focus on genuine emergencies.
A 2025 fire in the Vuache Tunnel in the French Alps demonstrates the importance of this technology. AID detected a stopped truck and alerted operators. Officials were then able to close the tunnel before the fire fully erupted. Thanks to the early warning, the driver was evacuated, and no other vehicles were involved.
Building a safety strategy
Combining complementary technologies lets agencies tailor detection to environments, monitor roads, proactively improve safety, enhance planning, and advance Vision Zero.
