Transformation principles

ankarsvik's picture
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Type: 

Compiler/Assembler: 

From time to time, I?m asked to explain how to transform from a coordinate system to another. Now when Sweden is converting from RT90 to a new planar system, SweRef99 TM, it?s time once again. It?s plenty of theoretical papers at www.lantmateriet.se for this subject.

I've cooked the essence of this into one Excel-file and one C-file.
The Excel-file seldom miscalculate more then a few centimetres.
The C-file adoption for AVR-processors can miscalculate a metre or so from time to time.
This is most because of limitations in SIN() and COS() functions in math.lib.
I added atanh directly in the mainfile.

The package contains a few PDF documents from www.lantmateriet.se .

Software is written and tested on an ATMega2561V and STK500.
It loops through 15 control points with a delay of 5 sec.
A single transformation takes approx 6 msec.
A 20x4 LCD display is also attached to PortC.

Update:

I went curious about the UTM international transformation.
The fact that Swedish "new" grid is based upon it..

However a search at the internet gave me the fealing that even profesors at Green Bay, US where members of "Flat-Earth-Society"....

I went to a dozen homepages, all members of F-E-S.
But then, I found Alan Morton's page http://www.dmap.co.uk/utmworld.htm

In main.c you'll find a few coords all calculated with utms.exe except the fact the prog is also a member of FES.

But still, please have fun!

Oops, today it's impossible to upload an extra zip-file....
This is an alternative link

http://www.ankarsvik.nu/download...

Plz be patient.

/Thomas J
Ankarsvik

Attachment(s):