Smart traffic signals are becoming a regular part of urban traffic management, helping to make roads safer for drivers, pedestrians and cyclists.
Smart traffic signals are becoming a regular part of urban traffic management, helping to make roads safer for drivers, pedestrians and cyclists.
Traffic signals are an integral part of keeping both drivers and pedestrians safe at intersections. As
traffic control systems have become more intelligent, the use of smart traffic signals to optimize urban traffic flow has become increasingly important.
For decades, intersections ran independently using inductive-loop traffic-detection technology. However, the advent of
Internet of Things (IoT) devices means more intelligent radar and
video detection sensors that can count, measure direction and speed of travel, and also determine whether an objects is a car, bike or pedestrian.
With this information, Todd Kreter, SVP and GM of Roadway Sensors at Iteris said, “traffic engineers at a central traffic management center (TMC) can immediately modify signal timing, including how long a particular approach gets red or green, and then optimize timing throughout the day as traffic volume fluctuates.”
More advanced smart strategies for traffic signals address highly dynamic changes in time durations for each and every movement within the intersection (cycle and split adjustments) and also across arterials (offset adjustments), explained Urban Traffic Management Experts from Kapsch TrafficCom.
Acquiring data for these adjustments can be done by using detectors (e.g., magnetic loops) or more advanced sensors, including video-based detection, and
radar/doppler technologies. Additionally, advanced strategies are not limited to
traffic counting, speed and occupancy measures. Adjusting cycle, split and offset in intersections in short intervals (e.g., 5 seconds) provides continued adaptation to varying traffic, and manages proactive adaptations, according to Kapsch TrafficCom.
Adaptive signaling regulation can be taken a step further with
artificial intelligence (AI), using rules-based engines, machine learning or other AI capabilities, including recurring situations and also “human behavior,” to solve the most difficult situations, Kapsch TrafficCom said.
Rapid Flow Technologies’ Surtrac traffic signal system combines concepts from AI and traffic theory. The system allocates green time to different approaches at intersections in real time to optimize the movement of actual traffic on the road.
“The system puts computing at the edge (i.e., a computer at every intersection) to produce ‘signal timing plans’ in real time, collects information on approaching traffic in real-time from independent sensing devices (e.g., video cameras, radar, etc.) mounted at the intersection, and depends on real-time communication between intersections to achieve network level coordination,” explained Stephen Smith, Co-Founder and Chief Scientist at Rapid Flow Technologies.
Sweco is developing Smart Traffic, a traffic light controller that utilizes data already available from traditional loop detectors along with new data sources like floating car data, cameras and radars. The data from its real-time and predictive traffic model is fused into a reliable image of the traffic on the level of individual vehicles, cyclists and pedestrians. Based on the predicted arrivals of traffic at the intersection, green phases are scheduled in advance optimizing both the duration as the sequence. Controlling traffic lights in this way results in reduced waiting times and CO2 emissions, according to Bas van der Bijl, Manager, and Stefan Hjort, ITS at Sweco. “In addition it is also possible to inform road drivers about the scheduled green phases, offering the possibility to adopt their arrival speed at the intersection in order to prevent unnecessary stops and increasing the comfort,” they said.
Matthew Trushinski,
Director, Marketing,
Miovision
Miovision offers a smart traffic signal platform called TrafficLink, which provides a range of solutions needed for a traffic team to collect, monitor and understand their traffic signals. The solutions include a managed cellular connection, and tools for signal monitoring, video streaming, maintenance alerts, as well as traffic data insights. Their SmartSense component brings traffic AI to the intersection, processing data gathered by its SmartSense 360 camera and enabling vehicle detection and ongoing studies of traffic, said Matthew Trushinski, Director of Marketing at Miovision.
In terms of solution implementation, there are many challenges when it comes to urban traffic signal control. One, according to Sweco, is finding the balance between optimal traffic light control and providing a reliable prediction of the future green phases to arriving traffic.
“The earlier drivers are informed about the signal changes, the harder it becomes to react to the actual traffic situation at the intersection,” Sweco said. They suggest using the latest sensor technology (e.g., intelligent cameras) in combination with predictive traffic models to make reliable predictions of the arrivals of traffic at an intersection for the next minute, making possible to optimize the traffic light control and to inform drivers about the scheduled green phases for the next minute.
Other challenges include pedestrian and bicycle detection. Effective traffic signaling in urban road networks must be able to distinguish different traffic modes (e.g., pedestrian, bicyclist, bus, passenger vehicle) and utilize this information in traffic signal control decisions, Smith explained.
“Most current commercial vehicle detection devices are not capable of simultaneously detecting vehicles and pedestrians, and the option of using additional detection hardware to enable pedestrian detection is often not an extra expense that cities are willing to bear,” Smith said. The situation, however, is changing with more commercial detection companies introducing detection hardware capable of integrated vehicle and pedestrian detection.