Intelligent Traffic Light System

[BOBBY]

Hey all,
Couple of years ago,when i was an engineering student I tried to develop an intelligent traffic control light system.
Back then I observed a lot of pollution at the traffic lights and the wastage of the energy resource gasoline.Green light was on even though there were no vehicles in that direction.This made me think and the result Intelligent traffic control system(Not completed).
My design outlook was something like this,
A Microcontroller which will use the input from the pressure sensors on the road some 100 to 500 mts away from the traffic light.Microcontroller will calculate the number of vehicles,load of vehicles and depending on that time for a particular direction is calculated as against the traditional fixed time control systems.everything was fine but i was unable to create usable pressure sensor.Readymade sensors were very costly.My question is any one has any idea for an alternative for the sensor or changes to the design outlook .

thanks
Bobby

[ash]

Hey Bobby,
I suppose those ready made sensors are expensive because you'd need to measure loads of up a couple of tonnes!

If you can exempt that variable, you can try using proximity sensors or laser sensors to detect the presence of vehicles. Also, you don't need to dig up the road to install these sensors, unlike the pressure ones.

[MaRo]

Can't it be like a camera that can recognize the pattern of the cars, pattern recognition technique to recognize the pressure of cars.

[KINETIC_JOULES]

How about something like they use on the railroads? I think, there is a censor that detects when the train has passed that leads to the mechanics in the arms that go across the road. I know I may not make any sense, but what I'm trying to say is maybe it would be just as simple as a censor that detects when the cars have passed and there is no more traffic flow in that direction.

[ash]

Originally Posted by KENETIC_JOULES

How about something like they use on the railroads? I think, there is a censor that detects when the train has passed that leads to the mechanics in the arms that go across the road. I know I may not make any sense, but what I'm trying to say is maybe it would be just as simple as a censor that detects when the cars have passed and there is no more traffic flow in that direction.

Yep, you can certainly use any type of sensor that can detect a presence of an object! Im not sure, but I believe the sensors you are talking about in railroads are infrared proximity sensors. They sort of emit infrared light (like the one in your TV remote control) out, and if there is an object in front of it, the light will be reflected back to a detector thats next to the emitter. Because infrared essentially travels at the speed of light, the detection sensitivity is very accurate.. which is suitable for fast moving objects like trains

Can't it be like a camera that can recognize the pattern of the cars, pattern recognition technique to recognize the pressure of cars.

Ah, that requires a lot of image processing techniques I believe. Its certainly viable though; CCTV cameras can be modified to detect moving objects as well as their speed. Worth a shot!

Your pressure comment reminded me of a scene in Italian Job, where Charlie and the team looked at the height of the ground clearance of the armoured trucks to figure out which one was carrying the gold

[KINETIC_JOULES]

Originally Posted by ash

Yep, you can certainly use any type of sensor that can detect a presence of an object! Im not sure, but I believe the sensors you are talking about in railroads are infrared proximity sensors. They sort of emit infrared light (like the one in your TV remote control) out, and if there is an object in front of it, the light will be reflected back to a detector thats next to the emitter. Because infrared essentially travels at the speed of light, the detection sensitivity is very accurate.. which is suitable for fast moving objects like trains

Yeah. . . that's what I thought they used too. ^_^

[BOBBY]

will image processing, streaming live vedio 14 hours a day, be cost effective?


what i thought then was, pressure sensors would be energy generating sensors(pressure of vehicles could generate power) so no need of external power to them as against others like infrared sensors and railroads are not as busy as normal roads and we come to know the time range and the number of trains before hand unlike vehicles on normal roads.May be we may use some kind of probability here on roads!

[ash]

Originally Posted by BOBBY

will image processing, streaming live vedio 14 hours a day, be cost effective?


what i thought then was, pressure sensors would be energy generating sensors(pressure of vehicles could generate power) so no need of external power to them as against others like infrared sensors and railroads are not as busy as normal roads and we come to know the time range and the number of trains before hand unlike vehicles on normal roads.May be we may use some kind of probability here on roads!

Yeah, thats what I thought initially. Piezoelectric pressure sensors would be suitable for that purpose, as they generate a tiny amount of voltage. But, you still need to amplify that signal so it can be used by other components in the system. Thus, the system cannot be entirely powered by those sensors alone.

It'll be hard to maintain a pressure sensor system, since you'll need to dig up the road a little (and again to replace damaged ones). You'd also need to lay the pressure pad across the whole road, since motorbikes could easily fit through any gaps.

You can always run infrared sensors off small solar panels. In fact, you can attach a radio module with it too, so data can be sent to the Traffic light 500m away without cables attached

[BOBBY]

But i feel a MODIFIED pressure sensor would be one time investment and nil maintaince kind of thing.
Design something mechanical device to overcome disadvantages of wornout of piezo electric materials?I too thought of solar energy Solar panels a bit costly in india and current traffic light system use solar energy in our city but i see them not operating on rainy and cloudy days.
I can see that solar system and radio module to it on bright days have definite advantage over other options.

[ash]

Originally Posted by BOBBY

But i feel a MODIFIED pressure sensor would be one time investment and nil maintaince kind of thing.
Design something mechanical device to overcome disadvantages of wornout of piezo electric materials?I too thought of solar energy Solar panels a bit costly in india and current traffic light system use solar energy in our city but i see them not operating on rainy and cloudy days.
I can see that solar system and radio module to it on bright days have definite advantage over other options.

Good point about the solar, however, you can always have a battery back up. Lead Acid ones can provide enough juice during the rainy and cloudy days Radio modems won't be much effected by rain attenuation as well, compared to WiFi. You don't need to buy a large solar array to operate sensors, unless they are ultrasonic ones. Infrared LEDs are quite efficient, and can be run off small PV panels.

Im not sure if this can be considered an advantage, but you can always relocate and scale up the number of infrared sensors since they are on the surface. For pressure sensors, you'd have to keep on digging and surface the road often if you need to expand or modify its location.

Don't get me wrong, I'm not totally against pressure sensors. However, its best to try to use sensors that are easier to implement, without disturbing much of the road surface.

[KINETIC_JOULES]

I think it should be some infrared censor that triggers the light to turn after a certain amount of time of no traffic.
Nuff said. . .