Flutterby™! : computerizing bike ride support 2010-08-02 22:32:17.956206+00

computerizing bike ride support

2010-08-02 22:32:17.956206+00 by Dan Lyke 5 comments

Re SAGging the Agony: One of the things that makes that ride special is that they've got a roughly 2 to 1 staff to rider ratio (normally is probably < 1:10), which means every rider is tracked pretty meticulously. This is a process that takes about 4 people per SAG station (you'd think it would be less, but in practice it was 4 of us sitting around a lot, and then working together to make sure the data really did get in all the places it needed to when that burst happened). I think it should take 1 and a computer, or be automated, and if that can be implemented then maybe it could be scaled from less than a hundred riders to thousands.

With that in mind, I'm interested in info on:

experience with long-range (20 miles or so) line-of-sight WiFi as an alternative to HAM packet radio.
getting started with HAM packet radio.
cheap long-range (5 foot or so) relatively high speed (< 30MPH) RFID transponders.
possibly solving that same problem with barcodes (IR lit up security cameras for scanners?)
old fashioned line-at-a-time dot matrix printers for data logging redundancy in harsh environments. One of the things I was thinking about was "what happens when the computers fail", and bomb-proof logging to human readable media seems necessary. But all the old MX-80s I've seen now fetch a premium, and I don't know where you'd buy new.

A few bytes a second would work. For the Agony, we could even just use 3G as there's at least AT&T coverage out there, but there's also the geek factor of building stuff that can be used independent of the big corporate network. I'll take a look at the Proxim stuff, I think we'd really need on the order of 15 miles or so, which makes going back to 2 meter and relatively high power transmitters a good idea again.

RFID-wise, I've only played with things that have ranges sub-inch. Multiple reads is no problem (I can guess at how frequently a bike should be going through gates), but I've yet to run into off-the-shelf antennas with range of several feet that still have reasonable transceiver costs.

And, yeah, my idea would be to build the system for the Agony, 'cause they've got the staffing to do backup human systems, test it there at their scales, and if that works then amp it up to a few thousand users and try to make a grand or two for a day in the sun. Not a living, but a hobby that pays for itself.

20 miles can be done with cheap WiFi gear (with the latency settings tweaked), and great antennas, but you may get better results with gear designed for those kinds of ranges, or rented sat-phones and/or boosted cell phones. The old 3-watt bag-phones with a good antenna will reach and receive farther, but suck at data and may or may not have freq coverage. Proxim makes some good long distance gear.. (or at least, did), The older Cisco Aironet stuff was higher power (low speed) and would do 5 miles easily.

How fast of a connection do you need? Would a solid stable 128-256kbps be enough?

As for RFID, it's all in picking out the right antennas and dealing with duplicate/multiple reads.

If you built a nice system, could you rent it out to the rides to cover some costs?

If, for a few bucks a rider, there were a way to track riders in and out of checkpoints on the larger rides, every ride in my area would buy it. And I think that if you gave riders a printout of leg times at the end of the ride, they'd cooperate with minor inconveniences (ie: ride through the gates). And double-century riders would totally cooperate, because they're usually working on credit for the California Triple Crown.

One of the problems around here is that we've got enough riders who aren't participating in any given ride that driving sweep is really hard (I've tried to do it on one leg of the Wine Country Century). And trying to get reasonable data out of rest stops can be tough ("Yeah, there's lots, get off my back so I can make sandwiches").

I'd guess that FourSquare wouldn't work on the Agony because of lack of cell phones (there's coverage in the valley, but not many of that demographic seem to carry cell phones yet), and, of course, in western Marin and Sonoma cell phone coverage is super sketch. Whence my questions about HAM packet radio; there's already APRS used on a lot of rides.

The Agony is special, but that could also make it a great testing ground. And even if the price was relatively high, double centuries would love a solid tracking system that didn't rely on hollering numbers and writing them down, or stopping at rest stops, and double riders would happily participate.

Aaand, tracking is already a big deal in running events, but that's a slightly easier problem (put an RFID tag on the shoe and antennas in the mat, you've got a reasonable time of stationary few inch proximity).

Ok... just read SAGging the Agony. There is definitely reason to track every rider there... But it's a small enough group, you may be able to just use FourSquare and have them check in at each aid station.

Hmmm... weekend before last, I was lead SAG for the Sunrise Century. They had about 800 riders. There's no cell coverage over most of the mountain part of the course. We had 4-8 SAG drivers at any one time- just going back and forth over the open sections of the course. One SAG driver was also the sweep.

The biggest problem we encountered was that our mechanics mostly dropped out the week before. But we had only a few adjustments and I don't think anyone had to drop out due to a failure. We did have a couple tires and the one blowout was a match.

I think it's easy to over-think SAG. The best thing the organizers did this year was drop the 50 and 65 mile courses. That meant only people who were up for at least 75 miles and a buttload of elevation gain (starts in Boulder at about 5400 feet - I was stationed at about 9500 feet).