News   Jul 12, 2024
 1.1K     0 
News   Jul 12, 2024
 977     1 
News   Jul 12, 2024
 367     0 

Roads: Traffic Signals

Hmm, i'm disappointed they went with SCATS. From what i've seen, InSync is way better. It does everything SCATS does and has cameras that monitor traffic flow and intelligently adjust the signals to it. I also believe InSync is cheaper. Weird.
Monitoring traffic flow and intelligently adjusting the signals is the entire purpose of an adaptive system. Every adaptive system does that, including SCATS and the previous SCOOT system.
 
Monitoring traffic flow and intelligently adjusting the signals is the entire purpose of an adaptive system. Every adaptive system does that, including SCATS and the previous SCOOT system.
The InSync system does it better than SCOOT or SCATS from what i've seen. Neither SCOOT nor SCATS use cameras to accurately detect and account for traffic volumes.

You can throw around buzz words like adaptive system, but to me any system that is just taking button pressing from pedestrians and in floor sensors etc is an outdated system. The systems that use actual cameras to monitor traffic flow are lightyears better.
 
The InSync system does it better than SCOOT or SCATS from what i've seen. Neither SCOOT nor SCATS use cameras to accurately detect and account for traffic volumes.
Cameras and inductive loop detectors are functionally identical. They provide exactly the same information to the system: the position, speed and direction of vehicles. Furthermore they are interchangeable. Any system can operate with either cameras or inductive loops. Even in Toronto we have intersections which use cameras and inductive loops for different directions at the same intersection.

In what way does InSync do a better job of detecting vehicles than the other systems?

You can throw around buzz words like adaptive system, but to me any system that is just taking button pressing from pedestrians and in floor sensors etc is an outdated system. The systems that use actual cameras to monitor traffic flow are lightyears better.

Adaptive Control is not a buzzword. It is the technical term which refers to a traffic signal control system which measures traffic flow over a period of time and determines a signal timing plan based on an optimization function in a model of tbe traffic system. In short it changes the maximum green durations based on the average traffic in the past 15 minutes or so.

Traffic Adaptation does not involve responding to individual road users, for example by holding the green for a group of approaching vehicles in real time. Real time adjustments occur via vehicle actuation. Adaptation and Actuation can be mixed and matched, because they operate on different timescales. A control system can be both adaptive and activated, or adaptive but not actuated, or actuated but not adaptive, etc.

An example of a buzzword is "smart traffic lights" which is a completely meaningless term which gets used for anything from basic vehicle actuation, traffic adaptation, any kind of communication, signal priority, etc.
 
Last edited:
Cameras and inductive loop detectors are functionally identical.
Absolutely not. An inductive loop detector can only tell that there is a car above the inductive loop detector. The cameras can count exactly how many cars are in the left turn lane, and literally see the cars leaving the lane and hold the light until they clear the intersection (within reason)

Cameras are functionally way better than any other system because they simply give you more data, end of story.
 
Absolutely not. An inductive loop detector can only tell that there is a car above the inductive loop detector. The cameras can count exactly how many cars are in the left turn lane, and literally see the cars leaving the lane and hold the light until they clear the intersection (within reason)

Cameras are functionally way better than any other system because they simply give you more data, end of story.
You really can't judge the end of the story without having read any other part of the book.

Inductive loops already do all those things which you are claiming to be unique features of cameras.

The time it takes to clear the intersection is also irrelevant to the duration of the green light (and thus also irrelevant for vehicle detection). That's what clearance time is for.

If I feel like it this evening I'll write a longer response to explain the basics of vehicle detection to you. In the meantime, check out this blog post of mine:
https://ontariotrafficman.wordpress.com/2020/10/10/how-vehicle-actuation-works/
 
Traffic signals are being installed on Shuter Street at Sackville and Sumach, replacing pedestrian crossovers.

I don’t think these are at all necessary, given the slow speeds on Shuter (it’s a two-lane 30 km/h zone, and there are protected bike lanes and street parking) here.

I am happy to throw the signal group of Transportation under the bus..........

However, they're off the hook here.

Have a look at what last year's report said:


1689106671859.png



It was the interim Councillor who imposed these with this motion:

1689106801360.png

 
This is usually how it goes. I often look up the approval process when I see a signal go in in a completely stupid place, and usually Transportation Services recommended that the signal not be installed but Community Council voted to overrule them and install it anway. From what I've seen, the staff in the signals operations group, District traffic operations groups and TTC operations group are all opposed to the ridiculous overproliferation of signals in illogical places, though the District often faces political pressure to adjust their recommendations to reduce opposition to the proposed unwarranted signal.

I think that we (the public) and City staff need to help educate council about the fact that signals are not some magic solution that makes traffic collisions disappear, and help them understand the roadway adjustments which actually do improve safety. On roads with 2+ lanes per direction, there are unfortunately few alternatives, but when there's only one lane per direction there are many safer and faster ways of getting pedestrians across the street.
 
Absolutely not. An inductive loop detector can only tell that there is a car above the inductive loop detector. The cameras can count exactly how many cars are in the left turn lane, and literally see the cars leaving the lane and hold the light until they clear the intersection (within reason)

Cameras are functionally way better than any other system because they simply give you more data, end of story.

Agreed. I was recently stuck at the stem of a T intersection for several minutes in a long line of cars because the car at the head of the left lane stopped one car length back, behind the loop. Eventually people started using the right lane to turn left on the red in front of him, and when our turn came we stopped, got his attention, and motioned for him to move up (he was on his phone without a care in the world). After we made our left on the red across him he finally moved, and in the rearview mirror the light cycled and began clearing the backlog. That would not have happened with a camera instead of the loop.
 
Agreed. I was recently stuck at the stem of a T intersection for several minutes in a long line of cars because the car at the head of the left lane stopped one car length back, behind the loop. Eventually people started using the right lane to turn left on the red in front of him, and when our turn came we stopped, got his attention, and motioned for him to move up (he was on his phone without a care in the world). After we made our left on the red across him he finally moved, and in the rearview mirror the light cycled and began clearing the backlog. That would not have happened with a camera instead of the loop.
It also would not have happened with proper in ground loop detector placement. Putting only a single loop in the lane guarantees lots of wasted time because the signal has no way of knowing what cars are coming until they are already entering the intersection, by which time it's too late. I explained this in my blog post linked earlier.

capture.jpg
 
Last edited:
It also would not have happened with proper in ground loop detector placement. Putting only a single loop in the lane guarantees lots of wasted time because the signal has no way of knowing what cars are coming until they are already entering the intersection, by which time it's too late. I explained this in my blog post linked earlier.

Very true, but I doubt that Waterloo Region would ever pay to put multiple loops into any intersection, especially this one which is slated to become a roundabout. (It's only had signals for two years.) A camera at least can be redeployed.
 
Very true, but I doubt that Waterloo Region would ever pay to put multiple loops into any intersection, especially this one which is slated to become a roundabout. (It's only had signals for two years.) A camera at least can be redeployed.
Indeed this is precisely the advantage of cameras: they are much cheaper to install per detection zone, since one camera can inclue many detection zones. With more and more cities moving towards cameras rather than in-ground detectors, they should update their detection zone configuration standards to take advantage of the fact that you can add additional detection zones at no cost (using the same camera). So far they seem to be replicating the old inefficient detection layout which was designed based on minimizing the number of detection zones to cut costs.
 

Back
Top