Below within quotes is my original post under the topic "RFID conundrum for start / finish line timing of mountain bike race". Instead of creating a new thread I have chosen to modify the title of this one.
I am now proposing,
An Arduino Uno at the start connected to an Xbee, ID-12 RFID reader and a photoelectric switch.
An Arduino Mega at the finish connected to an Xbee, 3.2inch TFT LCD module with touch panel & SD card cage, ITeadstudio ITDB02 Arduino Mega Shield V1.1 for LCD Module & DS1307 RTC, ID-12 RFID reader and a photoelectric switch.
At the start a rider would first scan an RFID card to identify themselves then they can pass through the photoelectric switch onto the course which will start the timer. At the finish the rider will first pass through a photoelectric switch and then scan their RFID card to register the time recorded as their own.
The Arduino at the start would only need to transmit the RFID card value and switch state to the Arduino at the finish.
The Arduino at the finish would be responsible for everything else...
Does this sound realistic or is it back to the drawing board again?
Thanks in advance for any input, Jason
Hi All,
I am currently working on a downhill mountain bike race timing system based on a Mega 2560 with a real time clock and an lcd display. The Ardunio will be located at the finish line and will be directly connected to an RFID reader and an Xbee. Connected to another Xbee will be another RFID reader located at the start line.
The problem is I dont know how to spec the RFID system for my application. Ideally the antenna would be contained in a 2m x 0.5m mat and the system would be able to read Tags upto 0.5m above the mat that are travelling at a velocity of up to 10 m/s. It would only have to read one Tag at any one time. All I have seen in relentless googling is complete turn-key timing systems for huge amounts of money.
Thank you in advance for advice or suggestions, it will be very much appreciated!
Ideally the antenna would be contained in a 2m x 0.5m mat and the system would be able to read Tags upto 0.5m above the mat that are travelling at a velocity of up to 10 m/s.
This is not the sort of thing you could do with a normal passive RFID system. The range is too large.
You might try and look at active tags, those have small batteries in them that last about three years. And yes they are an enormous amount of money.
Yes but championchip requires the tag to be on the shoe and therefore the read range need only be very small. http://en.wikipedia.org/wiki/ChampionChip
With a passive EM RFID tag you can only get a range of about 12" if that. That is what all the low cost RFID readers are. I used to design RFID readers for a living.
To get greater ranges with passive tags you have to go to UHF readers and they are very expensive.
Then you are certainly a good person to talk to and I think you for taking the time to respond.
As I know next to nothing about this, is it possible to use a battery activated passive tag with any low cost passive RFID reader or do they require specific hardware? Assume that a read rage of 12" (30 cm) is workable would it also be possible to read a chip that passes over the antenna at 10 m/s ?
If it is battery activated then it is not a passive tag, it is an active tag and we are back to those hands free readers.
The cheapest RFID tags operate at a speed such that it takes about 50mS to read a tag. At a speed of 10m/S the active read area has to be half a meter for the tag to be in the field long enough. That is way too much for a passive tag, I think you will struggle to do this with low cost technology.
The fact that there isn't a commercial system to do this (is there?) probably tells you there is not a simple way of doing it. How about a 2D bar code on the vest and a photograph at start and finish?
Well there are championchip, ipico etc but obviously they are shall we say, not affordable.
I would like an automated system and thought using RFID would be pretty simple and not that expensive. Machine vision was considered a possibility but deemed unsuitable as on a wet muddy course it would end up unreadable by the time the rider finished.
For (car) sprint racing we use a lighting split beam and have somebody read the number off the side of the car to say which car it was.
How much separation will there be between consecutive riders? If they are going to be multiple seconds apart, you might be able to get something working by using a simple 'bike presence sensor' (timing beam, pressure pad etc) to give you an accurate time, and a short range transmitter on each bike that just keeps broadcasting "I'm number seven" (or whatever) over a range of a few yards. If your bikers are going to be crossing the line close together I can't see that working, but if they're coming through separately you might be able to get something to work.
Alternatively, for multi-lap races we often use active transponders and a beacon - the hardware is quite expensive to buy but you may be able to rent it - people who have splashed out on this sort of thing may be keen to recoup some of their costs?
Please elaborate as your suggestion sounds entirely feasible.
It works like this, Riders start at the top of the hill and race down the course one at a time to the finish at the bottom. A really short course would take a fast rider approx 30 seconds to get down but a slow rider on the same course may take 90+ seconds. Rider starts are staggered by 30 second intervals so although it would be bad practice to send a slow rider down before a fast rider it is still feasible that there would be 2 or more riders on the course at any one time and they could finish in a different order than they initially set off. Riders would typically cross the line separately.
That's it in a nutshell.
Initially I had planned to wire a infra-red photoelectric switch into a motorola walkie talkie as the push to talk button to trigger the vibrate function in a receiving walkie talkie wired into an arduino and start a timer. At the finish would be another photoelectric switch connected directly to the arduino that would record the time. While this would work it seemed somewhat crude and a waste of the hardware, also it would only allow one rider on course. RFID seemed ideal for the project until I found out the cost was significantly more than I first expected!
Does this sound realistic or is it back to the drawing board again?
I guess I fail to see the solution here. Are the racers racing each other, or the clock? If they start at staggered times, more than one rider at a time may reach the finish line. Lining to to scan RFID tags to stop the clock hardly seems like a viable solution.
How is sending the switch state going to prove useful? What you need to send is the RFID tag and the time that the switch was triggered. That defines the start time for a particular rider.
The end time will be defined when the rider scans his/her tag again. Getting them to not pass each other on the course, and to line up neatly at the finish line will be the biggest challenge.
more than one rider at a time may reach the finish line. Lining to to scan RFID tags to stop the clock hardly seems like a viable solution.
That is a possibility but not that regular an occurrence for it to be a hindrance
How is sending the switch state going to prove useful? What you need to send is the RFID tag and the time that the switch was triggered
By switch state I just meant that it had been triggered.
The end time will be defined when the rider scans his/her tag again
That is correct
Getting them to not pass each other on the course, and to line up neatly at the finish line will be the biggest challenge.
If the system logs the time a tag scans onto the course and then the time it scans off the course then it would not matter if the riders were to pass each other of if a rider failed to ever finish, would it? Fundamentally it comes down to associating a time stamp with an tag.
Can you tell me if it is possible to connect a real time clock and an Xbee to an Arduino uno at the same time as they would both require the serial connection?
Can you tell me if it is possible to connect a real time clock and an Xbee to an Arduino uno at the same time as they would both require the serial connection?
Sure it is, and no they don't. I'm not aware of any RTC with a serial interface.
Sure it is, and no they don't. I'm not aware of any RTC with a serial interface.
Ah excellent, I must have crossed my wires somewhere while I have been trying to work all this out.
Aside from the points raised do you see anything else I have overlooked regarding this proposal? Can you think of a better way to accomplish what I am trying to do? I know I could use a UHF RFID reader and do away with the photo-switches completely but the cost is prohibitive.
If I have a real time clock module in the start Arduino and a real time clock module in the finish Arduino can I get them to automatically sync time with each other?
Also, am I correct in assuming that the Xbee is transparent to the Arduino so that I can prototype this all on a bench by connecting the boards via a serial cable and not have to cover the additional expense of the Xbee's until the software part is actually finished?
If I have a real time clock module in the start Arduino and a real time clock module in the finish Arduino can I get them to automatically sync time with each other?
No. On the other hand, you don't need to. Scan a tag. Send the data. The receiver time stamps the receipt of the tag, defining the start time. Scan a tag on the other end to record the end time for that tag.
What constraints are you trying to meet? What sort of distance will there be between the start/end points? How tolerant will you be of false/missing results? Does it need to be tamper-proof? Can you get away with simply banning overtaking? Is there anything stopping you having a human element involved in the normal operation of this?
