Can SX be used to detect vehicles
william chan
Posts: 1,326
More topics to discuss the capabilities of the SX :-
You know, those inductive coils embedded near the vehicle barrier gates to detect the passing of vehicles.....
Can the SX be used to detect the vehicles by connecting it directly to the coil?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
You know, those inductive coils embedded near the vehicle barrier gates to detect the passing of vehicles.....
Can the SX be used to detect the vehicles by connecting it directly to the coil?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
Comments
You could monitor·a freq, based on the L of the loop in an oscillator and assume that when it's outside some parameter that a vehicle is being detected (all that metal changes the loop's quiescent L.)· In real life they use a "loop amplifier" for this sort of thing.· I have a traffic background, so I know a little about this stuff, you know.·
[noparse][[/noparse] The loop's L can be affected by temperature, temperature change, and a host of other environmental and physical·factors. ]
http://renoae.com/traffic/index.php?option=com_content&task=view&id=107&Itemid=72
·
I have an interest in traffic control in the fact that with traffic analysis this country could save enormous amount of gas useage. What kills me first of all is in a simple situation of an intersection that continues to cycle through light changes whether there is traffic or not. Let's say there is only traffic from one direction, let's leave the light green until other cars are detected from the oposite direction. I have come up with the idea of traffic analysis in that a computer system and sensors actually monitor traffic conditions and also learns from traffic patterns and time. By doing this it also learns better flow conditions from other traffic lights. Now of coarse surrounding businesses wouldn't like this but tuff. I think our cities and goverment should spend some money into traffic analysis.
BigK
Post Edited (BigK) : 10/10/2007 2:12:34 AM GMT
To compensate for temperature, does the system use a differential loop coupling approach, or is it more of an adaptive approach (i.e. high pass filter - only responds to sudden changes, gradual changes are ignored)?
william chan,
There are several ways to sense the presence of metal (a Vehicle) within the proximity of a coil. Depending on what you are willing to invest may help determine what technique you use. My suggestion would be to go with something that is low power as possible so that your coil does not become a transmitter. Differential sensing is good, because you can negate any temperature changes but it requires careful positioning of 3 coils in order for it to work correctly.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
The coil would set up a an rf field tuned to a specific frequency and the passing steel undercarriage would affect the field strength.
Once again, to keep the frequency within bounds, a traditional fixed frequency sinewave oscillator [noparse][[/noparse]a tank circuit made of a coil and capacitor in parallel] would do the job best.
When you look at it, it requires merely two passive components - a coil and a capacitor. So why try to replicate it with a microprocessor. Even if you are concerned about frequency drift and have a crystal controlled oscillator, there still are not may parts involved.· Google 'Colpits oscillator' or 'Hartley oscillator' for some traditional standards.
On the sensor side, there is potential to have a triggered response converted into digital information. The first thing that comes to mind is measuring the vehicles speed from two or more sensors. Then there is the arbitration of cars coming from different directions and the count of vehicles backed up. Even more interesting is how the processor will respond to unexpected behavior - people running red lights and cars going the wrong direction across the sensors. Since autos have killed more people that all the wars in history, safety is likely to be the key factor in design.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Everything in the world is purchased by labour; and our passions are the only causes of labor." -- David·Hume (1711-76)········
Post Edited (Kramer) : 10/10/2007 9:20:09 AM GMT
BigK -- Ordinarily, the main street (non-acutated phase) should be held in green till the cross-street gets a detection within the "permissive" period.· That's basic fixed-time/pre-timed synchronization (co-ordination.)· That works best where intersections are regularly spaced and speed limits are constant across long runs.· Lots of times the loops get cut during road-work and the intersection is in constant re-call till the situation can be remedied (by cutting a new loop channel in the asphalt, which is a real big job.)· There is a lot of money being misspent on traffic analysis, already; truly, more than you know.· I'll bet (if you live in America) that your city is spending a lot of Federal grant money on an·Intelligent Traffic System·the central premise of which is "adaptive control."· The problem is that when the computer (a.k.a. "the system") is let to "manage" the situation (it is to laugh) -- people will start making phone calls (as they should.)
Kramer -- No RF fields here, mate, it's all low-freq stuff; people wouldn't·do with·their radio going squirrely on top of having to stop at a light.· The principle is basic, quiescent L vs. change produced, as you deduced, by a mass of metal in its proximity.··Commonly, a·small sign blank (aluminum sheet) is used to test with.
The technology is about high reliability.
If the coils were balanced, what would you expect to "see" on the output (secondary windings) if a sine wave were applied to the primary?
The answer would be almost no signal, because the opposite wound coils will cancel out each other.
Now suppose the inductive balance of the coils were thrown off by a piece of metal or an approaching vehicle, what would you expect?
The answer would be a signal proportional to the differential imbalance of the two secondary coils.
By detecting the Phase Polarity of the output compared to the input, you can also determine which direction the object is approaching/leaving the coils.
Edit
Here is a thought... supply the primary coil with a low voltage 60Hz signal.· Connect one side of the secondary coil·to a virtual GND (Vdd/2) through a voltage divider (two 1K resistors).· Connect the other end of the secondary coil to an I/O ...··Use the Comparator configuration,·or use the·TTL pin configuration to monitor the incoming pulses.
·
For example, configure the I/O pin·for TTL input
·
Note: optionally configure the TTL pin as a Comparator set to something other than 50% threshold ... (For this example use a setting less than 50% ; TTL is normally set at about 28%)
·
1) reset a counter
2) When the I/O pin goes LOW,·start a counter
3) When the·I/O·pin goes HIGH, then stop the counter
4) The value in counter should contain a value proportional to the differential displacement of the coil imbalance.
·
Note: For #4, the output signal from the secondary coil may need to be attenuated if the results are consistently 8.333 mS wide (1/2 Period of 60Hz) due to signal saturation.
········Simply place a voltage divider across the primary coil outputs, and connect the·output of the divider to the I/O configuration·mentioned above.
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
Post Edited (Beau Schwabe (Parallax)) : 10/11/2007 3:14:54 PM GMT
It seems to me that simply having traffic lights go off and on with a simple timed sequence which is what they seem to have is rediculous when the system can actually in the first simpliest form leave the light green in one direction for traffic flow if no other traffic is coming from the other direction. Second by controlling the lights with a computer(controller) you can also elliminate accidents from people running red lights, once the car is detected by speed or other means it now knows to hold the opposite traffic red light longer before changing. There are many reasons for my traffic analysis system idea. If the system is designed correctly those people that are generally on their phones anyway won't have to interupt·their call·to make another.
As it should be america needs to look at ways to save in gas useage and if it includes traffic analisys so be it, it's better to spend that money there instead of a wastefull war.
Just my opinion of coarse!
K
There are also pushbuttons at the traffic lights for the pedestrians to request a green light for the pedestrian crossings.
In some cities, you will also find a system called the "Dusseldorf funnel" (named this way because it was introduced in Dusseldorf first). These are special lights indicating the speed you should drive to approach the next crossing when the green light is on.
As you can see, many features can be implementd into a traffic analysis and control system. It requires a lot of experience to set it up right, and - in most cases - a loger test and optimization phase is necessary before it is working as expected in reality.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,
G
Since detecting inductance changes in the coil is somewhat difficult and unreliable, especially when the road floods,
I was thinking of using another method of detecting vehicles.
How about using 2 or 3 PING))) ultrasonic detectors to verify the that vehicle has passed before closing the barrier gate?
What is your opinion on the reliability.
Thanks.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
IMO, Beau's DLC comments are very valuable. DLC does not rely on incuctance changes of coils, but on the change of coupling of coils caused by metal parts brought close to the coils. So ice bears would not have a chance incfluencing this system (unless they are carrying a fridge, or something similar).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,
G
· There were several ultrasonic (overhead) detectors placed before I came on the job, with only one remaining when I showed up.· It worked well, actually, but it was placed up on a signal mast arm (appx 25 ft. up.)· For whatever reason, I guess it didn't prove itself in the marketplace.
· There are microwave-based (radar) presence detectors and video-based systems, too, but sooner or later detection failures and false-calls evince.· At present, nothing is·nearly as dependable as a loop detector.
· Are you figuring to design something for personal use?
· If you could design a scheme that combines high-reliability with low-cost installation --·you'd probably win the Nobel Peace prize, too.·
at my job, (public saftey), we use camera systems only when we can not cut up the roadway to install loops. Loops are by far the preferred method.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's Only A Stupid Question If You Have Not Googled It First!!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John J. Couture
San Diego Miramar College
The problem is that a pedestrian was lucky to make it to the island in the middle of the road in one cycle. For elderly, it was nearly impossible to cross the street. These lights were synchronized in the 1960s to 25mph flow, but less and less cross traffic access is allowed.
I suppose there is an obvious conclusion. All traffic control systems reach a limit a peak traffic densities where there is nothing to be gained. And pedestrians will just have to burrow under or fly over in any efficient system.
I like downtowns and communities that actually ban significant portions of vehicle traffic as people tend to respond to the better ambiance. It has been found that having computers limit the entry to freeways during peak periods gets everyone home faster than use of an unlimited entry system. So now you sit at a red light on the on ramp and wait your turn.
I also would mandate disbursement of tall office buildings rather than concentration in one district. San Francisco now is proposing a 300 floor building at the old East Bay Terminal. It would be better to provide 6 50 floor building in suburbs that currently suffer a 2 hour commute into the city.
If we really got it right, there would be little need for traffic lights.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Everything in the world is purchased by labour; and our passions are the only causes of labor." -- David·Hume (1711-76)········
Between 6:00AM and 9:00AM they go a hundred MPH to the city and between 4:00PM and 7:00PM a hundred MPH back.
I doubt there is much you can do to slow down the average citizen on their way to and from work other than petition for speed bumps or hire an officer to park his car with a radar gun. About the only action you could take is if the vehicle belonged to the city. I did this once with a School bus clocked at 45 MPH in a residential. I simply used a video camera running at 30 frames per second and timing/counting how many frames it took the bus to travel from two known distance reference points within the video. After about a week there was a new bus driver.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
···· I am thinking an sx28 on a shoe box bread board with a murata 50mhz resonator. Rs485 through a cat5 300+ feet along with power and develop a modbus protocol. Maybe·four reflective sensers, two up and two down the road a piece.
·· I wonder if anyone has used the transceiver set that parallax offers?
Nothing to offer to the thread, except to say that this is·very interesting stuff...
have never had an insight into traffic management like this before.
Our business is in fuel data recording systems for fuel dispensers, and we are always looking at new ways of identifying vehicles when they pull up at the pump, hence my interest in the vehicle detection.
Cheers,
Chris, Western Australia.
You'll have better luck starting a new thread with your own topic rather than adding a post to the end of a two year old thread.
Thanks,
PeterM