Where are the trackpoints in SBN files?

> 10 years ago
Reply
Register to post, see what you've read, and subscribe to topics.
Dylan72
Dylan72
QLD
683 posts
QLD, 683 posts
21 Jan 2009 11:22pm
Since everyone is insisting that GT-11s have to have their Alpha 500 results calculated using trackpoints, I've been investigating what is involved, and I've discovered something very interesting.

It seems that there is no trackpoint data in the SBN file format. (Spec is at http://www.navmanwirelessoem.com/oem/customer-support/oem-news/product-briefs-and-data-sheets/jupiter-32-xlp-new2/sirf-binary-protocol-reference-manual and the relevant message ID is 41, on page 87)

For each point, there is only one speed measurement, and it is a doppler one.

So I am a bit confused, because obviously Realspeed and GPSResults seem able to work out a "trackpoint" calculation for Alphas without actually having any trackpoint speeds. They must be derived somehow from the other data, perhaps working backwards from the locations?

Any idea how?

I'd love to understand this a bit better.
decrepit
decrepit
WA
12884 posts
WA, 12884 posts
21 Jan 2009 10:47pm
Dylan not sure of my ground here, but isn't the data between two consecutive track points a distance measurement, at a fixed time apart. Think this is how trackpoint speed is derived, distance over time.
nebbian
nebbian
WA
6277 posts
WA, 6277 posts
21 Jan 2009 11:07pm
It's bits 24-27 (Latitude in degrees (+ = North) x 10 ^ 7 ), and bits 28-31(Longitude in degrees (+ = East) x 10^7) isn't it?

Page 3-42 (the one after you said), Message ID 41.

There's a big table that shows which bits do what.


Well done on finding that document btw, when I was thinking of doing something similar I searched high and low but couldn't find it anywhere.
Dylan72
Dylan72
QLD
683 posts
QLD, 683 posts
22 Jan 2009 12:14am
Hmm, when I originally put together my first tentative code for this project, I assumed that the lat and long points were what all the speed and distance calculations were based on. That line of thought got abandoned early on, after the results I got were well off the results in Realspeed.

Once I do a bit of testing, I'll give this approach a go, but if anyone has any better ideas in the meantime, please let me know. To do the speed calcs using the points could make things run a bit slower, I am using some fancy trigonometry for the job.

Ta,
Dylan.

nebbian
nebbian
WA
6277 posts
WA, 6277 posts
21 Jan 2009 11:27pm
I'm sure you could find the speed between successive trackpoints, and create an array based on that.

Then your fancy trig algorithms would still work, and you'd be doing it on trackpoint data
Dylan72
Dylan72
QLD
683 posts
QLD, 683 posts
22 Jan 2009 7:38am
nebbian said...


Well done on finding that document btw, when I was thinking of doing something similar I searched high and low but couldn't find it anywhere.


Well, you can thank the guys who work so hard on GPSBabel for that. There was actually a link to it in the comments of the source code for the SBN file code.

I think if it wasn't for the GPSBabel project, this sort of thing would be a lot trickier.
whippingboy
whippingboy
WA
1104 posts
WA, 1104 posts
22 Jan 2009 12:09pm
Careful mate you might open the can of worms that is GPS data software.

There is no way I would rely on these GPS units software for anything requiring scientific type accuracy. Think of those missiles that go slightly off target

However, as the Challenge is not for sheep stations and everyone uses the same software (I'm talking in the actual GPS's here) they are satisfactory for comparative purposes.

"The accuracy of GPS technology is relative, not absolute"
mathew
mathew
QLD
2172 posts
QLD, 2172 posts
22 Jan 2009 1:33pm
whippingboy said...
"The accuracy of GPS technology is relative, not absolute"


Could you clarify what this statement means -> in the literal sense, lat/lon is an absolute position (which includes some position error).
whippingboy
whippingboy
WA
1104 posts
WA, 1104 posts
22 Jan 2009 3:03pm
I think Ive even confused myself

What I mean is that your position is determined relative to the location of the satellites.

Your GPS then calculates your position in lat/lon, but then again all measurement is relative and relies on the accuracy of instrument calibration.

Now I'm stuck in a philosophical paradox loop
My brain hurts

Resume normal transmission
Please Register, or first...
Topics Subscribe Reply