Repository navigation
Coordinate conversion from ECEF to ECI does not match pymap3d #3
Description
Try finding the coordinates of Null Island (geodetic latitude 0, longitude 0, elevation 0) at noon on January 1, 2000 as per the following code:
[ecef_x, ecef_y, ecef_z] = geodetic2ecef(0, 0, 0)
J2K0 = datetime(2000, 1, 1, 12, 0, 'TimeZone', 'UTCLeapSeconds')
[eci_x, eci_y, eci_z] = ecef2eci(ecef_x, ecef_y, ecef_z, J2K0)
Then run the same code via pymap3d and you will see the difference:
[J2K0_whole, J2k0_fraction] = py.matlab_datenum_to_datetime(J2K0)
J2K0_for_pymap3d = J2K0_whole + J2K0_fraction
eci_vector = py.pymap3d.ecef2eci(ecef_x, ecef_y, ecef_z, J2K0_for_pymap3d )
where matlab_datenum_to_datetime explicitly transforms the date as follows:
function [whole_days, fractional_days] = matlab_datenum_to_datetime(matlab_decimal_datenum):
PYTHON_TO_MATLAB_CORRECTION_CONSTANT = td(days=366)
whole_days = [dt.fromordinal(int(date)) for date in matlab_decimal_datenum]
remainder_days = [number%1 for number in matlab_decimal_datenum]
fractional_days_uncorrected = [td(days=remainder) for remainder in remainder_days]
fractional_days = []
for day in range(len(fractional_days_uncorrected)):
next_fractional_uncorrected_day = fractional_days_uncorrected[day]
fractional_days.append(next_fractional_uncorrected_day - PYTHON_TO_MATLAB_CORRECTION_CONSTANT)
end
I cannot figure out where the error is coming from, so I resorted to calling pymap3d from MATLAB since it uses astropy. You might want to get rid of this repo and just recommend that route if you cannot fix the issue.
Activity
scivision commentedon Sep 2, 2020
This is expected from the simple ECI transform in this repo that does not account for nutation and other factors. Putting a position on the equator of a spheroid without those higher fidelity models in Astropy results in ECI also being on the equator.
I am assuming the Astropy transformations are non-trivial, but I didn't investigate.
The Matlab Aerospace Blockset has ecef2eci() if needed or yes use Python + Astropy.
fPkX6F1nGTX commentedon Sep 3, 2020
I edited the MATLAB code above so it has fewer bugs. I do not have access to a MATLAB license so I cannot verify what it returns, but I did run the following Python code just now, for reference:
The magnitude of the vector returned is very, very close to 6378137, the X-coordinate of Null Island in ECEF.
fPkX6F1nGTX commentedon Sep 3, 2020
@scivision If you could run the MATLAB code and report what it returns for Null Island's ECI coordinates on January 1, 2000 at noon, that would be greatly appreciated. The lat/long/alt is 0, 0, 0 and ECEF is 6378137, 0, 0; I would also appreciate knowing the difference between the
ecef2ecianswer and thegeodetic2ecianswer, if any (would depend whether lla2ecef returns the coordinates I mention above, as per https://planetcalc.com/7721/ (input 0 latitude to get radius of Earth at equator).