Week 3 - Smart Signals

As technology has relentlessly progressed to newer heights, many technologies need improvement, and they risk becoming obsolete. One of these technologies is traffic signals; however, changing ALL of America’s aging traffic signals is a monumental task. Local and state DOTs should reconsider certain intersections and turn them into roundabouts or focus on other forms of transport, as I talked about in my last blog. However, many major intersections are constantly inundated with car traffic with no end in sight; the upgrades that I will discuss should be applied to them, along with making room for public and active transport.

Smart traffic signaling aims to optimize traffic flow at signalized intersections. The method of achieving this objective is to decentralize the process of signal timing and phases, using detectors at intersections. The detectors can be cameras, radars, or radio frequency devices to collect traffic volume, speed, and congestion level data. With these devices, it means that the traffic signals can make split-second decisions to respond to changes in traffic flows.

The decentralized nature of the signals makes them independently optimized. However, it is also possible to allow for signals to communicate with each other at different intersections. This lets the intersections create a ‘green wave,’ a corridor where motorists hit all green lights. The corridor with the most traffic is given green lights to aleve congestion.

So how do these newer, ‘smart’ algorithms work?

One example of smart signaling technology is Scalable Urban Traffic Control (SURTRAC), created by two Carnegie Mellon professors, which uses proprietary software integrated with existing traffic control devices for optimization. The professors did share a technical overview of the ideas that govern the operation of the SURTRAC system. Initially, “the traffic at each intersection is analyzed as a mathematical problem known as a ‘machine scheduling problem’ where traffic flows are represented as aggregates or sequences of queues of vehicles for signal allocation” ¹. These schedules are then used to determine whether there should be any phase changes. Such as continuing to give a green light or turning red so that traffic on other lanes can be let go. This control is called the ‘Schedule Driven Intersection Control’, which shares its information with the ‘Schedule Driven Coordinator’. The coordinator shares its schedule with neighboring intersections, who then input it into their own calculations for systematic optimization.

This is a system similar to other forms of systems that were created to solve the waning signal problem, such as the Adaptive Control Software (ACS), Real Time Traffic Adaptive Signal Control System (RTTRACS), and Automatic Traffic Surveillance and Control (ATSAC) ². These were developed, in part, by governmental bodies such as the Federal Highway Administration. These systems have a more centralized structure connecting them to a traffic operations center. However, it results in latency which will affect the ability to respond to real-time conditions. There have also been many new applications with AI that can even more effectively make schedules and decisions based on even rudimentary data, having the ability to improve phase scheduling.

Detecting pedestrians is equally important in adjusting signal operations, and the British created their own pedestrian detection system. The pedestrian user-friendly intelligent (PUFFIN) crossing uses active and passive detection; a pedestrian walks up to the crossing and presses the push button to cross. The “controller checks the curbside presence detector. If a pedestrian is present, the pedestrian phase is called, and a light is illuminated, informing the pedestrian of the call” ². Then, the pedestrian phase is called, and pedestrians are allowed to cross; the advantage of the system is then shown, with the phase continuing until the pedestrian is detected on the other side of the crossing. There is also the added benefit that if the pedestrian crosses when it’s not their phase or leaves the detection area, the pedestrian phase is canceled, reducing delays for cars.

Now that I have learnt about many different types of smart signals that are applicable for both motorists and pedestrians, next week I will take a dive into the costs associated with these signals and how cities in the US can purchase them. This signaling technology should be paired with pedestrian friendly-designs to ensure walkability and safety.

Here are my sources for the blog:

¹Smith, Stephen, et al. Non-Market Strategy Analysis Project Report, 2014. https://www.cmu.edu/epp/people/faculty/course-reports/SURTRAC%20Final%20Report.pdf
‌
²Fong, Gene, et al. Signalized Intersection Safety in Europe, December 2003. https://international.fhwa.dot.gov/pubs/pl03020/pl03020.pdf

That’s all for today!