Slashdot videos: Now with more Slashdot!
Until four years ago, I worked in Los Angeles as a traffic engineer. The ATSAC system is used on more than 4,000 intersections, is interconnected, and makes adjustments to signal timing either manually or automatically. There aren't cameras at all of the monitored intersections, but you don't need cameras to measure traffic volumes and speed, they're just an additional tool.
Further, the systems that use cameras for vehicle detection are falling out of favor. There are too many conditions, such as rain, snow, fog and bright sunshine that can befuddle the systems and cause them to fall back to pre-set timing rather than relying on vehicle detection. Where inductive loops aren't used, radar is proving to be more reliable than cameras.
Even where there's no interconnection, most intersections have controllers that are considerably more sophisticated than simple timers.
But consider cameras used not to give tickets but to adjust light timings in real time.
It's already being done. Los Angeles started in 1984 in anticipation of the Olympics, with system called ATSAC. There are several different types in use today.
Three seconds is the minimum duration as per federal law.
What gave you that impression? Signal timing is determined by an equation, but may have some minimum defined by local or state law, or in a design manual.
At 25 MPH with no grade, depending on law or policy, it may be perfectly permissible to have a 2.8s interval, though it's common in practice to round that to a minimum 3s. Here's the equation from the 1999 ITE handbook:
Y=t + 1.467v/2(a+32g), where
t=perception-reaction time, typically assumed to be 1 s,
v=speed in MPH (ITE recommends using the 85th percentile speed, but many agencies use the posted speed.),
a=deceleration rate, assumed to be 10 ft/sec^2,
If you want to read more, NCHRP report 731details recommended guidelines, and provides some history for the recommendations.
Thanks. Why wouldn't radar work? Mountainous roads?
Ethanol, by itself, has an octane rating of about 129. The octane rating isn't about the energy content of the fuel, but rather its tendency to ignite under compression ("detonate") as compared to iso-octane.
Those tube-shaped sensors can probably measure a bit more than speed. I don't know what is used where you are, but I recently completed the design for a dozen automatic vehicle classification stations. They measure speed, count axles, and of course, count vehicles. I doubt that the sensors you're talking about are for speed enforcement: That's easily done with radar and photos.
In this context, they're talking about a test sample. Perhaps the summary should have been edited to say so in plainer English.
"What if I suddenly feel like driving from Florida to Alaska?"
How about Arizona to Fairbanks. Seventeen days for the trip is quite a long time, though. I've driven between LA and Fairbanks twice, and can do it in 7 days without pushing too hard.