Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

Few ideas not yet mentioned for gyros:

• A gyroscopic presentation mouse could possibly be mounted in a fixed position. The receiver is USB, so that's how it would have to talk to the PDA. ($70 USD)
• A nav system with a built-in gyro -- normally only expensive nav systems have it, but geeks.com has an exceptionally cheap one. The down side is you have hack two devices; the Maemo app/driver, and the GPS unit. ($150 USD)
• A bluetooth gyroscope for aviation. This device must certainly have the precision required, but only practical for millionaires. ($1k USD)

Also, does the application you folks are discussing require OBD updates more frequently than ~0.03Hz? If so, that could be an issue. This video review of Carman/Elmscan 5 shows very sluggish realtime performance from the OBD interface (guages update once or twice a minute).

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

If that is the case, then it would probably make more sense to have the system log both GPS (phase 1) and OBD-II (phase 2) data, then run the analysis offline. We will know more once a developer comes forward to let us know which features are viable and which are not. :)

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

Several things (note I'm working on doing Zmapper and already get bus data from my Harley (If you have google earth, try http://homepage.mac.com/tz1/.Public/HarleyTelem2.kmz)

Not the track, but my Gear and RPM:

http://homepage.mac.com/tz1/.Public/rpm%2Bgear.jpg

This is with a 5Hz GPS, and not fully interpolating the data.

But let me break some misconceptions and myths.

The location is ACCURATE to only a 2.5-3 meters, but is PRECISE to a much smaller space especially over a very short time when the satellites used aren't changing and all have the same error. Otherwise every fix would swing in and out within the 3 meters. It doesn't. It often moves by a micro degree or two at most.

Also, the speed measurement (actually you can get XYZ velocity vectors on some) is also quite accurate, which you would need for performance.

Accelerometers may help but how accurate (and calibrated) are they? You don't need much for a Wii using it as a game device, but it isn't likely to be precise enough for doing performance work. The chips inside might be capable, but you might need to worry about noise, temperature, etc. if you want it to differentiate between 0.65 and 0.66g instead of just 0.6 and 0.7.

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

Agreed. Hence, the objective now is to keep the project simple and focus on the key functions. Having accelerometers is nice, but not critical to a motorsport application at this level. People who want a full-fledged tool for really high-quality data analysis are better off with a dedicated device such as a Racelogic Performance Box (http://www.performancebox.co.uk).

I also believe that it should not be impossible to have the application use the output file format from PerformanceBox, so that the tablet can be used as a portable analytics tool -- it sure beats having to keep a laptop in the car, or leave it in the pits during every track session

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

I have a n800 and an OBD2 scanner on its way in the post. I can help with testing if needed.

I'm a developer but I haven't done anything on the Maemo platform yet so not much use there at the moment but I can help someone else out if required.

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

I had developed pyAxelWii for this purpose of measuring accelaration (0 - 100kph), quarter mile time and also for checking the performance of my motorcycle after various tinkering. One problem I faced is that all the accelaration data should be captured fast otherwise, the end result may change.

I have currently postponed my development since my N800 has been sent for replacement and I expect to recieve it by mid march.

Currently I have a 50$ Bluetooth gps (holux m1000) which I am using to plot my position/speed on a webserver. I have a small board running debian http://www.wireless-products.dk/Lexcom/WP-CV862A.htm

I also plan to interface my car with OBD interface to get more info.

I would be very much interested in development in this area.

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

What kind of mounting do you have to keep the N800 on the handlebars?

if you have gpsd and python, here is the python code to get the gps data

import gps, os, time
session = gps.gps()
while 1:
os.system('clear')
session.query('admos')
# a = altitude, d = date/time, m=mode,
# o=postion/fix, s=status, y=satellites
print
print ' GPS reading'
print '----------------------------------------'
print 'latitude ' , session.fix.latitude
print 'longitude ' , session.fix.longitude
print 'time utc ' , session.utc, session.fix.time
print 'altitude ' , session.fix.altitude
print 'eph ' , session.fix.eph
print 'epv ' , session.fix.epv
print 'ept ' , session.fix.ept
print 'speed ' , session.fix.speed
print 'climb ' , session.fix.climb

time.sleep(1)

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

Wow, super simple. Didn't realise it was so easy to get gps data. I feel a coding session coming on for me to have a play with this somemore. :)

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

That's very useful to know. The other day I was criticizing most of the nav systems on the market for not having a gyro. (Aside from pricey dash mounts, only tomtom and cobra seems to have intertial nav as a backup). Someone told me the gyro is useless, because of the mutual gross inaccuracy of both GPS and maps. My claim was that with a gyro, a nav system would know to reroute quicker in the event of a deviation. Currently, most nav systems are unaware when someone deviates until the deviation is very significant, and then often the rerouting isn't complete until you're already at the next intersection. But according to what you're saying w/ precision being acceptable, information from a gyro would indeed enable a nav package to begin the rerouting process sooner. Correct? I'm also thinking a gyro could be used as sole navigation for tunnels and other GPS dark spots.

OTOH, if the precision of GPS is good, then why does it take so long for a GPS system like TTN6 to discover that I've deviated?

Re: Motorsport Performance Application - looking for developer to bring idea to reali
 

let's keep in mind that the intention here is to build a cheap and affordable GPS log analytics tool for the trackday/karting enthusiast. Most motorsport enthusiasts would not be overly concerned about their lateral G readings as they would be over stuff like average speed, sector time, lap time, etc.

Also, most tracks/circuits are outdoors with very few of them having coverage blackout areas, and even if there were, the small anomaly in reading would not be significant enough to affect the data much. Also, motorsport enthusiasts on a budget would not be overly concerned about having accuracy down past the 1/100th of a second, and if they did, I would happily recommend that they start considering a prosumer tool like the Performance Box that costs around 3 times as much.

The tablet's advantage here is that it is portable and has good battery life. If I could find someone to port EcuFlash (www.openecu.org), Mitsulogger (www.aktivematrix.com), and ScanTech (www.aktivematrix.com), and run it alongside the GPS log analytics tool, that would be perfect. Ecuflash would allow me to maintain my street and track ECU ROMs in my tablet and allow me to change maps at the track without needing to carry around my notebook. Likewise, I would be able to use Mitsulogger to do ECU datalogging from the OBD-II port, and use ScanTech to identify and remove DTCs and troubleshoot where needed.

Adding a gyro and other peripherals at this point in time would increase the complexity of development significantly without any huge gains.

For now, what I really need are suggestions on how to get freelance developers who are looking to put their name on a project like this, to get this project off the ground.

As it were, I'm almost prepared to run a small coding contest with a small prize pot of USD5000 to the winning application.

vabgeo, you seem knowledgeable on such matters. Are you keen to try your hand at coding this app?