As traffic signals needed to become more efficient and allow more traffic though an intersection, the industry realized that pre-set timed lights alone would never be good enough, even with different timings per day. The only way to streamline the process was to know which vehicle movement would best serve the current demand.
Because putting a traffic engineer on every corner wasn’t a possibility, controllers needed to have more smarts to make the best decisions. To make smarter decisions about how to run the lights in the intersection, controllers needed to know about the vehicles at the intersection. To solve this, in the early ‘60s, induction loops were deployed as detectors.
Induction loops use a simple interaction of metal, magnetism and electricity to detect large metal objects. They have many advantages: they are simple, cheap and reliable. However, they have some serious drawbacks.
Sometimes, loops can miss smaller vehicles, like motorcycles or bicycles. In addition, as modern cars get smaller and consist of more plastic than metal, they become harder to detect. Finally, loops are expensive to install (or reinstall), because they are embedded in the road. This leaves them vulnerable to unrelated road work, which can damage them or leave them inoperable.
Enter the Age of Video Cameras
While research using video detection cameras was done as early as the ‘70s, use of video detection systems was not widely spread until cheap, reliable cameras became available in the ‘90s and 2000s.
Video cameras solve many problems that loop detectors face. Because video detection is not based on the presence of metal mass, smaller vehicles can be detected. Cameras are also easier and less expensive to install and maintain, because they attach to mast arms, poles or nearby street lights.
Video detection systems provide an extra added benefit: they give operators the ability to see what is really going on at the intersection in real time. However, they brought along new classes of problems too like sun glare, fog and other weather conditions that can make vehicles detection difficult.
Other modern traffic detection options range from thermal cameras to radar, as well as other advanced sensors. All of these options provide a more accurate understanding of real-time traffic needs that help run an intersection more efficiently. Each category of detector has their own strengths and weaknesses. However, one deficiency has never been addressed by any of these detectors since loops were first deployed.
Loops can only detect presence or not, meaning they can only tell a controller if a vehicle is above them, or not. This is good information, but it isn’t enough information to run the intersection in the most efficient manner. To know how to serve the greatest number of vehicles, the controller needs to know how many vehicle requests are at each approach. Loops can answer: “Yes someone is waiting,” but they can’t easily tell how many. So how can a controller know which phase serves the longest queue of waiting vehicles?
Adaptive Traffic Control with Video Processing
In|Sync does special video processing, looking at the images of the cars waiting on each approach to calculate which movement best serves the waiting vehicles. It then places calls to the controller to serve those phases.
In|Sync continuously analyzes queue levels and wait times, placing calls to the controller for the appropriate phases. Using this method, In|Sync streamlines intersection operation, even with the same controller running the lights. In|Sync uses cameras, other modern detectors, historical data, as well as loops to detect, verify and serve actual traffic demand.
All detectors have weaknesses, but unless you can afford to place a professional traffic light operator at each busy intersection to assess traffic and run the signals, you will have to rely on automatic detectors of some kind.
Knowing this, why wouldn’t you prefer a more advanced system that can understand demand better and put it to use when trying to move traffic safely and efficiently though the intersection?
Finally, all of this focuses on a single intersection. For busy corridors, traffic engineers must consider coordinating signal timing between each intersection. That’s something In|Sync excels at as well but, that’s a topic for another time.
