GPS Update Rate

Robert T

Is there a GPS module which updates more frequent than 1 Hz?

I would like to build a device to record various parameters(acceleration, lean angle, speed, etc) when mounted on a motorcycle on a race track and then map these values to the track itself. With an update rate of only 1 Hz that would mean a GPS value only every 150 feet if traveling at around 100 mph and speed typically go as high as 160 mph.

max72

I had the same problem, and I'm testing the sanav FV-M8.

www.sanav.com/gps_engine_board/FV-M8.htm

You can change the baud rate and the uptade of the single strings up to 5Hz. Documentation is poor, but I was able to set baud rate ad update rate, playing with resets during the setup. I have a test demo updating up to 5Hz and saving to a SD, using existing code from obex. If you are interested in this GPS model I can upload the code, even if it is a work in progress, poorly commented.

Massimo

Robert T

Thats the best update rate I have seen so far. At 5Hz and 100mph, that still an update every 30 feet. For cornering telemetry I need and accuracy of 1 to 2 feet. That would mean an update rate of 50 + Hz. This may not be possible with commercial GPS modules.

Thanks for the info Massimo

Robert

SRLM

Sparkfun has a few. I've used this one, and wrote a Propeller object for it (or any other 5Hz GPS): www.sparkfun.com/commerce/product_info.php?products_id=8975

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Powered by enthusiasm

sylvie369

I think that most people would tell you that GPS is not the best source of data for the kinds of variables you're collecting, let alone for that kind of update rate. 5 Hz is the fastest I've seen as well, where even a simple, cheap accelerometer will easily do many thousands of times that rate.

Robert T

Sprakfun also has a GPS Module with a 10 Hz update rate.

http://www.sparkfun.com/commerce/product_info.php?products_id=9060


Sylvie369, you are probably right, It seems that more accurate modules are available with higher update rates but the cost gets much higher. A combination of techniques to calculate position may be the best method.

Robert

achilles03

Robert T said...
Thats the best update rate I have seen so far. At 5Hz and 100mph, that still an update every 30 feet. For cornering telemetry I need and accuracy of 1 to 2 feet. That would mean an update rate of 50 + Hz. This may not be possible with commercial GPS modules.

You won't get an accuracy of 1 to 2 feet out of GPS, regardless of your update rate. Even at 1000 Hz, you'll still be limited to about +/- 30ft due to day-to-day drift, reading-to-reading noise, etc.

Dave

Robert T

I must plead ignorant when it comes to GPS technology. Technical specifications for GPS module on the Sparkfun site claims an accuracy of 2.5m but a velocity accuracy of .1 m/sec. These numbers don't compute at the stated update rate. Similar numbers are given for the Polstar GPS module sold by Parallax.

I assume that the GPS units sold for running have similar accuracies. If thats true then a 30 foot error would mean that instantaneous speed calculations are not even close to being accurate.

Does the accuracy number mean that the difference between two consecutive readings separated by 1 second in a GPS module with a 1 Hz update rate may vary by up to the specified number(2.5m for example) or is the interval between reading longer or does it have to do with the difference between two reading separated by a specified distance regardless of time interval.

Robert

achilles03

Robert T said...
I must plead ignorant when it comes to GPS technology. Technical specifications for GPS module on the Sparkfun site claims an accuracy of 2.5m but a velocity accuracy of .1 m/sec. These numbers don't compute at the stated update rate. Similar numbers are given for the Polstar GPS module sold by Parallax.

I assume that the GPS units sold for running have similar accuracies. If thats true then a 30 foot error would mean that instantaneous speed calculations are not even close to being accurate.

Does the accuracy number mean that the difference between two consecutive readings separated by 1 second in a GPS module with a 1 Hz update rate may vary by up to the specified number(2.5m for example) or is the interval between reading longer or does it have to do with the difference between two reading separated by a specified distance regardless of time interval.

The accuracy SHOULD be the difference between where you really are vs. what the GPS tells you. The accuracy is typically around 30 ft or so. Between consecutive readings, or even readings over a few minutes, the difference will be less (maybe 5 to 10 feet over several readings), and will depend on signal strength, velocity, acceleration, etc.

Hope that helps,
Dave

Robert T

Thanks, that helps. I ordered a 1HZ GPS module last night from Parallax to experiment with.

Robert

hover1

Take a look at some survey GPS's. Do a Google search on that. The prices might be a shock. 10mm accuracy IIRC. I used to have a link to the system that NASCAR uses, but I can't find it right now. I think it was DGPS.

Jim

wjsteele

Precision and accuracy with GPS is very difficult. In order to get the best solution, it must be combined with Accelerometers and Gyroscopes to deal with the changes between the GPS updates.

The best thing to do is use a kalman filter to process the data. GPS is great over time, where Accels and Gyros are great for small distances. Combining the three together (maybe even with Magnetometers) will give you a better understanding of what is happening.

Bill