In 1912, Lester Wire, a young police officer in Salt Lake City, came up with an idea. What if there was a tool to regulate cars at intersections instead of patrol officers, who needed to spend hours rooted to a platform through rain, heat, and hail?
Wire came up with a wooden box on a pole. It had two light bulbs inside, colored red and green. The box was connected to electricity so the light bulbs could be switched from one to the other with the press of a button. That’s something patrol officers could do from a booth at the side of the road.
Since then, traffic light signals have evolved a bit. We now have yellow and don’t need a patrol officer to press a button. But the original concept has remained largely the same — traffic lights change on a pre-programmed schedule.
However, the state of our roads in the twenty-first century is much different than it was 100 years ago. We have more cars, bigger road networks, higher population densities, and constant traffic disruptions.
Perhaps it’s time to rethink the old and introduce a smarter traffic light system.
What is a smart traffic light?
A smart traffic light is an internet-connected vehicle traffic control system capable of adapting traffic light controls based on information collected from sensors, edge devices, and video systems.
At the intersection, smart traffic lights look the same as regular traffic lights except for extra hardware elements such as IoT sensors and/or connected CCTV cameras. On the back end, smart traffic light systems are connected to a cloud-based traffic management platform. They are often powered by predictive algorithms for dynamically adjusting traffic signals.
A quick disclaimer before we go any further: A smart traffic light system can’t miraculously fix all road issues, such as congestion, accidents, and rule violations. But they are a better preventive measure than traditional traffic lights.
As Dan Saffer, an author and the Creative Director at Smart Design, says:
Traffic lights are only a mechanical prop, a signifier of a social contract we’ve agreed to (and have written into law).
Apart from a potential fine (and good conscience), nothing stops people from red-light running (RLR) on empty intersections — and drivers do that a lot. In New York City, more red-light violations were recorded in 2021 than in any year since 2014. Accident rates also went up, which is problematic.
Why do people violate traffic signal rules?
Scientists agree that violations are highly contextual. The exact reasons vary, but they often fall into one of these categories:
- Intersection layout and rules. The type of intersection, signal countdown timers, signal mounting configurations, and signal timing are some of the factors that can prompt drivers to bend the rules (when no one’s watching).
- Road user behavior. When drivers see others breaking the rules, they’re inclined to follow suit. Seeing a preceding vehicle or a vehicle in an adjacent lane passing through the intersection on yellow is strongly associated with RLR.
- Variable circumstances. The time of day, day of the week, and weather conditions also affect RLR rates. People are more likely to engage in RLR in the mornings between 06:00 and 12:00 — in other words, when they’re likely in a rush or running late.
Smart traffic light systems cannot fully discourage people from breaking traffic rules, but they can make it less tempting.
With adaptive traffic signal control (ATSC), you can program dynamic rules for signal changes based on conditions and better detect RLR at busy intersections. Smart traffic signs can also adjust recommended speed limits based on the weather or road conditions to improve traffic throughput. Intelligent traffic lights, in turn, can adjust signal timing based on the volume of vehicles at different intersections and variable factors such as the time of day. Such a setup can ensure smooth traffic flows and reduce the number of situations when breaking traffic rules seems appealing (or undetectable).
For city managers, intelligent new traffic lights are a much-needed alternative to manual or rule-based signal controls.
Many cities operate traffic management centers that are hives of activity akin to a busy urban air traffic control operation. But appearances are somewhat deceiving, given that the traffic engineers have limited tools available to manipulate their signal networks to respond in real-time.
Urban traffic managers can (and should) integrate smart signaling into an intelligent transportation system. Merging the two enables users to exercise algorithmic, context-driven control over the city’s transport grid through one interface. The best part? Digital traffic signals can be dynamically adjusted in real time across the entire network to:
- Give priority to public transport and improve scheduling
- Provide special controls for emergency services vehicles
- Instruct last-mile commercial fleets to optimize delivery routes
- Alleviate signs of congestion at the onset
That’s some blissful city to live in, right?
Sadly, only a fraction of new traffic signals today are smart. Traffic light hardware can last for up to 30 years if well-maintained. But among new traffic signals, few are (or can be) connected to cameras, radar systems, or sensors. And those that do have basic sensing capabilities often can’t detect cyclists or pedestrians.
Fortunately, this is changing. Urban planners — and the general public — realize that standard traffic light systems need extra wits as more connected cars and electric commercial fleets hit the roads.
As part of a £30m investment in better urban connectivity, the city of Manchester is testing an AI traffic lights system. The city is installing a network of edge devices for collecting real-time road data on each junction and plans to use 5G technology for dispatching data to the cloud for analysis.
London, in partnership with Siemens, is testing a real-time adaptive traffic signals control (ATSC) solution called Sitraffic FUSION that is powered by data from connected vehicles and connected road infrastructure. Sitraffic FUSION can detect, model, and optimize routes for all modes of transport around the set KPIs. The system also includes a traffic light algorithm for optimizing controls on signalized junctions and pedestrian crossings.
Many more cities are looking into modernizing their traffic controls — and that means plenty of opportunities for new market entrants.
How do smart traffic lights work?
Smart traffic signals are equipped with sensing, video capture, and connectivity technologies to collect real-time data from the environment. The obtained data is either pre-processed on the device or transmitted to a cloud-based transport management system, where it’s processed by a predictive traffic light algorithm that generates instructions for signal adjustments.
A standard smart traffic light system has two elements:
- Roadside unit
- Cloud control center
Smart traffic light hardware
A smart roadside traffic light unit still has the familiar three-light interface — and some extra goodies.
- Connectivity modules — Wi-Fi, 4G/5G, V2X, GPS
- Modern traffic lights must be capable of collecting and exchanging data with connected cars, onboard vehicle computers, telematics systems, cloud-based traffic platforms, and mobile travel or driving apps.
- Sensors — variable specifications
- Radar/LiDAR
- Speed detection
- Weather sensors
- Emissions capture sensors
- Connected cameras with real-time traffic detection capabilities
- Red-light running monitoring
- Pedestrian and traffic flow detection
- Vehicle detection for traffic counts
- Accident detection
- Onboard computer — variable specifications
- Dynamic smart signals require an edge device with sufficient processing power to pre-process captured traffic data and execute adaptive controls.
The exact configuration differs by manufacturer. Some smart traffic lights have more advanced sensing capabilities; others just rely on camera footage. NoTraffic, for example, uses IoT sensors that rely on radar and computer vision for smart signaling and also captures car data in C-V2X and DSRC formats.
Cloud control center
On the software side, a smart traffic light system can process data in two ways: on-device (on the edge) or in a cloud location.
On-edge road data pre-processing reduces latency. With suitable hardware, you can run baseline traffic conditions analysis on a smart traffic light device. For example, such roadside units can:
- Analyze vehicle movements at intersections to detect violations
- Count vehicles to adjust signal timing
- Estimate emissions levels to inform urban planners
On-edge processing is a pillar for implementing adaptive traffic signals control (ATSC) — real-time traffic signal adjustments based on the current road situation. ATSC systems can reduce average travel times by 25%, shorten signal wait times by 40%, and lower emissions by 20% according to Carnegie Mellon University.
Next, digital traffic signals can dispatch pre-processed and raw data to a connected cloud-based control center, such as an intelligent transport system (ITS). Here you can perform more advanced modeling and predictive analysis to stave off traffic congestion and harmonize public transport schedules.
Likewise, you can use historical data collected by edge devices to build advanced models for:
- Traffic prediction
- Parking management
- Multimodal transportation planning
- Mobility as a service (MaaS) solutions
Must-have features for a smart traffic light system
Smart traffic light technology adds a new dimension of real-time control — and many good things come as a result:
- Reduced travel times. The average American spends 58 hours per year waiting at traffic lights. Smarter controls can make people and goods move faster through city arteries.
- Less pollution. Idling cars emit 30 million tons of CO2 into the atmosphere every year. Reduced wait times and fewer traffic jams translate to cleaner air.
- Fewer road accidents. About 90% of road accidents happen primarily due to traffic violations. Smarter traffic light systems can minimize the temptation to bend the rules.
- Higher public transport ridership. Prioritized signals for public transportation can make public transport more attractive. In New York City, for example, a 15-minute shorter commute translates to 25% higher rail service usage.
To provide the above benefits to urban planners, future traffic lights should include the following four features.
Adaptive traffic signals control (ATSC) for urban traffic
Predictive algorithms operating at the back end of a smart traffic light system can find effective solutions to complex traffic management problems. Such systems can correlate traffic signaling rules with violation or accident rates — and model risk-minimizing scenarios with higher precision than a human traffic manager could (plus do so in real time).
Over time, a predictive smart traffic lights system can rely on sensors and visual data alone to make on-the-spot decisions and control traffic movements.
Case in point: A group of German researchers recently collaborated with city planners in Lemgo on an AI program for traffic light management. The team used a set of high-resolution cameras and radar sensors to capture traffic data. Then they trained a deep learning algorithm to regulate signaling at a busy intersection.
The algorithm was tasked with estimating the optimal switching behavior for the traffic lights and the best phase sequence to reduce:
- Waiting times at the intersection
- Average journey times in the area
- Noise and CO2 emissions
During the simulation run, the algorithm managed to achieve a 10% to 15% improvement in traffic throughput in the tested area.
Integrated Emergency Vehicle Signal Preemption (EVSP)
Emergency vehicles need priority access to the roads. The chance of survival is reduced by 7% to 10% for every minute emergency medical assistance is delayed. Likewise, the consequences of delayed arrival of police, firefighters, and other emergency services can be grave.
Yet emergency vehicles often get stuck in heavy traffic where drivers have to move aside to let the emergency go through. A signaling system with a smart emergency vehicle traffic light changer can address the matter in four ways:
- Update signaling to help emergency vehicles move faster
- Change the grid signals to divert vehicles from an affected area
- Implement prioritized signaling near emergency vehicle garages, parking lots, or stations
- Inform drivers in advance about a passing emergency vehicle to give them extra time for maneuvering
Case in point: A recent US study of smart traffic signal preemption for emergency vehicles found some interesting insights.
- In Fairfax County, Virginia, a preemption system enabled emergency vehicles to pass busy areas faster and with fewer conflicts, saving 30 to 45 seconds per intersection.
- In the city of Plano, a similar solution reduced the average number of emergency vehicle intersection crashes from 2.3 per year to less than one every five years.
- The city of Plano also managed to maintain the same response times with fewer fire and EMS stations in the area.
Eco-driving mode
Vrooming engines at busy intersections create a layer of noise and air pollution. Plus, they make areas with busy intersections less desirable for urban dwellers — a factor that also affects a neighborhood’s economic development.
Smart traffic signals can help reduce vehicle idle time and promote more sustainable driving habits. A group of Taiwanese scientists carefully documented a set of eco-driving traffic light regulation models that can be implemented in sensor-based traffic light systems.
| Eco-Driving Model | Concept | Actions | Benefit | Application |
|---|---|---|---|---|
| MaxTM, MinADM |
Max, throughput and Min. acceleration and deceleration | OBU suggest eco-driving speed | Reduce carbon emissions, fuel consumption, travel time | Standalone intersection |
| VANATE-based coordinated signal control model | Forecasting and decision making | RSU determines traffic signal plan | Reduce carbon emissions and fuel consumption | Multiple intersections |
| EDAS System | Calculate number of intersections that can be passed and can different modes | OBU suggest eco-driving speed | Reduce carbon emissions, fuel consumption, travel time | Multiple intersections |
| TTA&RS | Travel time prediction and path recommendations | OBU forecast travel time and suggest path | Reduce computational complexity and reduce travel time | Multiple intersections |
Source: MDPI — Design and Implementation of a Smart Traffic Signal Control System for Smart City Applications
Their findings have already been put into action by the PTV Group in Taipei. The PTV Balance software platform can detect changes in traffic patterns and suggest smart stop light signals for cars, cyclists, and pedestrians.
Taipei authorities tested the platform in two districts, Neihu and Nangang. The results were impressive:
- 7.9% overall improvement in average travel time
- 12.6% reduction in travel delays on weekdays and public holidays
- 318,269 liters per year in fuel savings
- 101.1mt/year reduction in CO emissions and 720.2 mt/year reduction in CO2 emissions
Micro-mobility priority service
Micromobility vehicles such as bikes, e-bikes, and e-scooters are a growing part of the MaaS ecosystem. But they also present extra road hazards, both for pedestrians and drivers. In the first half of 2021, e-scooter crashes in London grew by 2,800% compared to the entirety of 2018.
As the use of personal and shared micromobility solutions surges, their movements must be better regulated. Smart traffic light systems should factor in these road players and create better controls for them.
For maximum safety, it’s best to adopt a two-step mechanism:
- Detect and recognize vehicles via smart traffic light video systems for traffic signal adjustments
- Alert nearby drivers of micromobility riders using V2X-based updates or signal phase and timing (SPaT) messaging
Peek Traffic has developed an interesting smart mobility solution for regulating cyclists and pedestrians They aim to connect all road users to an intelligent traffic control system via an ITS app. The app, in turn, sends signals to a smart traffic light control system. So when a vulnerable pedestrian attempts to cross the street, the app can issue an update to a connected traffic light so that it automatically adjusts the signal lengths. The same app can also inform the traffic light system about approaching cyclists to adjust the timing for them. Such dynamic traffic lights on cycling roads make bike use more attractive, which carries lots of benefits.
Smart Signaling for Safer Mobility
The impact of smart traffic light systems extends beyond mere driver convenience; it can create a cascade of positive effects:
- Prioritized signaling can make multimodal transportation more attractive and increase public transport use and profitability.
- Less traffic congestion and pollution paired with greater road safety makes certain neighborhoods more attractive for new residents.
- Prioritized emergency vehicle signal preemption increases the speed and efficiency of urban service providers (and reduces their operating costs).
- Integration of micromobility regulations increases safety on the roads and encourages more people to leave their cars behind.
- Finally, less congestion makes commercial deliveries faster, offering economic benefits for the area.
Given that our current traffic control systems are over a century old, updates are overdue.
Intellias is a technology partner to leading companies in the transportation sector. Contact us to receive a personalized consultation on developing smart traffic light software.
