PDA

View Full Version : UHF RFID lap counter project


pinwc4
2012.12.28, 10:14 PM
For quite some time I have been on the look out for a good RFID system to use as a lap counter. Finally I think I have found a UHF system that could be a good choice. My testing so far has been going very well. I put a small video together showing the lap counter in action.

Flip Side Lap Counter Video (http://youtu.be/gJjW4lKaLN0)

In the video I maxed out at 4 tiles wide but the system can go further depending on conditions. If the lap counter is mounted above the track instead of next to it you can get much wider coverage. Also the RFID tag choice impacts the range, I have tested 3 types so far and have more on order to continue testing. The tag I am using is similar to a credit card in size, but thinner and lighter, similar to the Core tags. I have more tags on order to experiment with as I think I can use smaller ones with no problem.

UHF RFID has a few advantages over the other RFID systems.

Adjustable signal strength, can do up to 4 tiles wide in tests so far
Better tag collision code, much less likely to drop laps because of multiple tags
900 MHz should not conflict with radios
Uses EPC Gen2 RFID tags which is a global standard


One main disadvantage I see so far is that the existing RFID tags people use will not work, neither Core or Kyosho. But RFID tags are cheap, less than 2$ each in small quantities, so this is not a big deal. This change is necessary though to get a technology that can be adjusted for different track widths. I tried various designs using the 13.56Mhz readers like Core and Kyosho use but never came up with something reliable for accommodating wider tracks.

If you did not watch the video above here is quick picture of the lap counter. It is the small white box to the right of the finish line.

http://www.flipsideracing.org/uploads/pinwc4/fsalpha1.jpg

I still have more testing to do but everything looks very promising so far.

imxlr8ed
2012.12.29, 12:23 AM
I'd be curious how it would pick up Xponders (tags) if you had one on the front of the car and one on the back, just to see if it would pick them up in that order consistantly. Also, since the signal reader seems to operate almost like a beam, could it be focused or condensed onto the area you designate as the finish line with shielding... something like a box to put it in, kinda like blinders on a horse?

Looks good so far! Best of luck!

LED
2012.12.29, 04:39 AM
And what would be the price of the system itself?

pinwc4
2012.12.29, 09:51 PM
I'd be curious how it would pick up Xponders (tags) if you had one on the front of the car and one on the back, just to see if it would pick them up in that order consistantly. Also, since the signal reader seems to operate almost like a beam, could it be focused or condensed onto the area you designate as the finish line with shielding... something like a box to put it in, kinda like blinders on a horse?

Looks good so far! Best of luck!

Good idea, I had already planned on running them side by side to see how that goes. I have ran it with one in the car and one on top the car with success already. Though I needed about an inch between them for it to reliably detect both.

To block some of the beam you could use something like chicken wire. Enough signal would be blocked by a wire mesh with holes no bigger than about 3 cm. Something like aluminum foil could work as well, though I will need to find the equations since thickness of the foil as it may be too thin to do with just one layer. Though really what I was doing was just offsetting the lap counter some so that where the detection starts was at the edge of the finish line.

But we would not want to block too much signal. The RFID tags themselves do not have any power source, they get their power from the signal sent by the lap counter.

And what would be the price of the system itself?

My guess at this point is if I were to do this it would be around $450 including 10 or so tags(transponders) and an USB to Serial convertor with a 5 meter cable. I think if I do sell this I will always include the convertor so I do not have to deal with some of the bad ones that people buy not knowing any better. Additional tags would probably be $1-$2 each.

saddad
2012.12.30, 05:08 AM
would the easiest thing be to mount it on the inside of the track facing out (oppiste where it it is now?), also could it mount ovet the track bridge style as giroz etc do now? but looks promising if the price is right

arch2b
2012.12.30, 10:10 AM
very promising!

very interested to understand, as ed is pointing to, how you can focus the detection range to a confined area so you don't pick up before a designated point and from adjacent lanes.

$450 is on par with all the other systems so i don't think that is a concern at all. $2 tags is a great thing. this is where other IR systems get you, with the transponders.

can you share some pictures of the tags your using in relation to say, a core tag? even core tags can get tricky with the complex rear end suspensions of today.

pinwc4
2012.12.30, 10:35 AM
would the easiest thing be to mount it on the inside of the track facing out (oppiste where it it is now?), also could it mount ovet the track bridge style as giroz etc do now? but looks promising if the price is right

You could mount it on the other side but you would have to shield the back side of the lap counter. The antenna picks up from the back and front. Plus you still have to route the data and power cables back.

On top would probably be a better choice than on the inside of the track. It has a pipe mount that can be used for this.

pinwc4
2012.12.30, 11:07 AM
very promising!

very interested to understand, as ed is pointing to, how you can focus the detection range to a confined area so you don't pick up before a designated point and from adjacent lanes.

With the way I had it mounted the detection was fairly uniform, better than a particular IR system I had used (not I-Lap). The main thing that will need to be done is to offset the lap counter past the finish line so that where the cars get detected lines up with. After more experimenting I will have a specific recommendation for installing.


$450 is on par with all the other systems so i don't think that is a concern at all. $2 tags is a great thing. this is where other IR systems get you, with the transponders.

That is part of why I like the RFID system. I prefer everyone be able to get their own transponder instead of having to share. The IR transponders understandably cost more as they are complex circuits. A good IR system has some advantages but I like the lighter weight and lower cost of tags along with not needing to power them from the car.


can you share some pictures of the tags your using in relation to say, a core tag? even core tags can get tricky with the complex rear end suspensions of today.

Not yet, I purposefully did not include a picture of the tag because I doubt the ones I am currently using are what I will recommend to people. It is very similar in size to the Core tag but I want something smaller. I have another sample pack on order that should arrive this week.

Based on the datasheet I think that the Alien ALN-9662 will be the standard tag. At 70mm x 17mm It is about 6mm shorter than a Core tag and 28mm narrower. At that size I think the tag can be installed so that the length of the tag crosses the width of the body, that way it will not be anywhere near the suspension or front tires.

arch2b
2012.12.30, 11:43 AM
there are the very small but finiky tags that worked for core. while core size tags are not a deal breaker, the smaller one can go the better. our club also favored the rfid system that did not require a connection to the car.
http://mini-zracer.com/mini-zgallery/data/1321/medium/IMG_07851.JPG
http://mini-zracer.com/mini-zgallery/data/1321/medium/IMG_07831.JPG

pinwc4
2012.12.30, 12:44 PM
there are the very small but finiky tags that worked for core. while core size tags are not a deal breaker, the smaller one can go the better. our club also favored the rfid system that did not require a connection to the car.


Good point, this also reminds me of the difficulty with mounting them to F1's.

I do plan experimenting more with tags, the sample pack has a variety of sizes including something similar to your second picture at about 2" x 2" and some about 2 3/4" x 11/16". I am still looking for a supplier for something smaller to try.

I expect I will stock a few different sizes for certain scenario's. So for example I plan on coming up with a good choice for dNaNo's. Since the track width is typically smaller we can get away with smaller tags. The smaller the tag, the less the range. So while with a Core sized tag at maximum power I can reach out up to 15 feet with something dNaNo' sized it may only be 3 feet max.

I have to come up with good compromises between tag size and detection range. But people will also be welcome to get tags from wherever, if I sell this I will not lock it down so that you must get them from me.

arch2b
2012.12.30, 01:06 PM
open source so to speak :)

can you tell us what the current tags you are reviewing are like? paper with sensitive coils embedded, plastic with internal coil? are they delicate or more robust than the core tags?

can the tags store data or does the computer store data with the UID? just wondering if they are universal or need to be configured at each location for a user.

pole mounting the sensor over a track doesn't appear to be a problem. it's not that big or distracting. may need to be painted black or if they offer black as an option?

reliability will always be the key question. we loved core but it was not as reliable as needed for competition. giro-z is better but still has it's own bugs. very much looking forward to the development of this system!

TheSteve
2012.12.30, 07:46 PM
Any chance of building a custom loop antenna that would go under the track, I guess it would be something like the Core one, may possibly be a bit more uniform across the track for counting.

Also, if you make an open source system you should be fine, if you put something into production be careful with existing patents on RF ID lap counting.

pinwc4
2012.12.30, 08:08 PM
open source so to speak :)

can you tell us what the current tags you are reviewing are like? paper with sensitive coils embedded, plastic with internal coil? are they delicate or more robust than the core tags?


The current tag I am using is a paper tag with an adhesive back. The ones I have coming in are thin plastic with the antenna coil, very similar in material to the generic tags you pictured.


can the tags store data or does the computer store data with the UID? just wondering if they are universal or need to be configured at each location for a user.


The support for this is the same as other lap counters, the computer stores the information associated with the UID. The tags do not have enough memory to store all the details that can be configured for a racer. I would rather not store any info on the tag anyways because if I do then the car must be at the lap counter any time a change is made to the racers information.

More than likely in the future I will have a web service that stores racer information to handle this. I have been getting more experience with creating web services and this seems like a decent approach. This will make it independent of the lap counter type. Though some systems such as Giro-Z will have to be excluded since they do not have unique id's per racer. The big drawback is the computer will have to have an internet connection to use the functionality.


pole mounting the sensor over a track doesn't appear to be a problem. it's not that big or distracting. may need to be painted black or if they offer black as an option?


It is about 10" square in size. It may take paint well, seems to be pretty standard plastic casing so something like Krylon should work. The manufacturer does not have it in other colors, and the quantities I would be getting are too small to justify a custom order.


reliability will always be the key question. we loved core but it was not as reliable as needed for competition. giro-z is better but still has it's own bugs. very much looking forward to the development of this system!

Yep, this is where I think it will do better than the Core. When there are multiple tags in the field at the same time I can detect each of them. I plan on testing the limitations of this though, going to mount a bunch of tags to my brushless rc18 to see how well it does.....

With the Core I could only detect 1 tag at a time. If multiple tags were in the field I would have to silence the first one detect to detect another. Or sometimes when multiple were in the field at the same time I could not detect any, they would not respond at all. Nothing I could do about it, it was a limitation of the hardware.

pinwc4
2012.12.30, 08:29 PM
Any chance of building a custom loop antenna that would go under the track, I guess it would be something like the Core one, may possibly be a bit more uniform across the track for counting.

Also, if you make an open source system you should be fine, if you put something into production be careful with existing patents on RF ID lap counting.

No, UHF RFID systems do not use an antenna loop like the HF (Core Speedway, Kyosho) systems do.

This is not something I have to put in production, I already found a supplier for them with a discounted rate for higher quantities. This is an existing product I am just configuring to be optimal for rc car lap counting.

Patents, and the absurdity that follow them, are always a concern. Really something like this should not be able to be patented, this is using RFID technology for what it was designed to do, track things. The whole purpose of RFID is to track packages going down a belt, or boxes moving through a warehouse, parking passes for getting into parking garages, etc. etc. etc.

But of course logic does not necessarily apply, I was very surprised when Core got a patent on their product which really is just applying RFID for what it was intended. However their patent does not apply to this solution for a few reasons so I am safe. I have had discussions with a patent attorney in the past about this and I will need to talk with him again.

But... even if I am in the right people are still free to sue anyone. Nothing I can do to prevent that.

saddad
2013.10.19, 11:00 AM
Any more news on this?

pinwc4
2013.10.19, 09:10 PM
Unfortunately not much news. A little bit after these posts our club lost access to the facility we used for racing so I never got to fully test the system. We got bumped for tumbling classes.

I use it occasionally at home with 2-3 cars on tracks in my garage or living room and it works great for that but it never has been tested in a large environment. From the behavior I have seen I have no reason to suspect it would have problems but just no longer have a way to test.

stevesteve
2014.11.04, 01:36 AM
Hi,

I would be very interested in this system. Did you do any more testing? Any progress in between? Is it maybe even readily available now?

If you would like assistance in testing, I would be willing to help and validate the concept in my tests.

Hope to hear from you.

Regards,
Steve