cysgp4 0.3.6

Project Status

Purpose

The cysgp4 package is a Cython-powered wrapper of the sgp4lib library (by Daniel Warner) to compute satellite positions from two-line elements (TLE).

It provides similar functionality as the well-known sgp4 Python package (by Brandon Rhodes), which uses Numba internally to speed-up the calculations. In contrast to sgp4, cysgp4 can work well with arrays of TLEs and/or times and make use of multi-core platforms (via OpenMP) to boost processing times a lot.

Installation

We highly recommend to use cysgp4 with the Anaconda Python distribution, in which case installiation is as easy as

conda install -c conda-forge cysgp4

Otherwise, you should install cysgp4 via pip:

python -m pip install cysgp4

The installation is also possible from source. Detailed installation instructions can be found in the user manual.

Usage

Using cysgp4 is possible via an object-oriented interface or with a fast numpy-array functional approach. The former works like this:

import cysgp4

# Define a date/time and an observer
pydt = cysgp4.PyDateTime.from_mjd(58805.57)
lon_deg, lat_deg = 6.88375, 50.525
alt_km = 0.366
obs = cysgp4.PyObserver(lon_deg, lat_deg, alt_km)

# Define satellite properties/orbit via two-line element string (TLE)
hst_tle = cysgp4.PyTle(
    'HST',
    '1 20580U 90037B   19321.38711875  .00000471  00000-0  17700-4 0  9991',
    '2 20580  28.4699 288.8102 0002495 321.7771 171.5855 15.09299865423838',
    )

# Create a satellite object for querying coordinates
sat = cysgp4.Satellite(hst_tle, obs, pydt)
sat.eci_pos().loc  # ECI cartesian position, km
(5879.5931344459295, 1545.7455647032068, 3287.4155452595)
sat.eci_pos().vel  # ECI cartesian velocity, km/s
(-1.8205895517672226, 7.374044252723081, -0.20697960810978586)
sat.geo_pos()  # geographic (geodetic) position, lon/lat/alt
<PyCoordGeodetic: 112.2146d, 28.5509d, 538.0186km>
sat.topo_pos()  # topocentric position, az/el/dist/dist_rate
<PyCoordTopocentric: 60.2453d, -35.6844d, 8314.5683km, 3.5087km/s>

# One can change time to determine positions at another moment
sat.mjd += 1 / 720.  # one minute later
sat.topo_pos()
<PyCoordTopocentric: 54.8446d, -38.2749d, 8734.9195km, 3.4885km/s>

In many cases, however, one probably wants to calculate coordinates for a (large) number of satellites, observer locations, and/or observing times. For this, the function ~cysgp4.propagate_many is useful. This is an array interface to the sgp4 calculations, which allows to perform calculations for different satellite TLEs, observers and times in a parallelized manner. ~numpy broadcasting rules apply:

import requests
import numpy as np
from cysgp4 import PyTle, PyObserver, propagate_many

# Download many TLEs from a website
url = 'http://celestrak.com/NORAD/elements/science.txt'
ctrak_science = requests.get(url)
all_lines = ctrak_science.text.split('\\r\\n')

# Need to convert them to a list of tuples (each tuple consisting
# of the three TLE strings)
tle_list = list(zip(*tuple(
    all_lines[idx::3] for idx in range(3)
    )))
# Create an array of PyTle and PyObserver objects, and MJDs
tles = np.array([
    PyTle(*tle) for tle in tle_list
    ])[np.newaxis, np.newaxis, :20]  # use first 20 TLEs
observers = np.array([
    PyObserver(6.88375, 50.525, 0.366),
    PyObserver(16.88375, 50.525, 0.366),
    ])[np.newaxis, :, np.newaxis]
mjds = np.linspace(
    58805.5, 58806.5, 1000  # 1000 time steps
    )[:, np.newaxis, np.newaxis]

# The result is a dictionary
result = propagate_many(mjds, tles, observers)
print(result.keys())
dict_keys(['eci_pos', 'eci_vel', 'geo', 'topo'])

# Returned array shapes are as follows; last array dimension
# contains the coordinate pairs.
print(np.broadcast(mjds, tles, observers).shape)
(1000, 2, 20)
print(result['eci_pos'].shape, result['topo'].shape)
(1000, 2, 20, 3) (1000, 2, 20, 4)

# One can also skip over coordinate frames.
result = propagate_many(
    mjds, tles, observers,
    do_eci_pos=False, do_eci_vel=False, do_geo=False, do_topo=True
    )
print(result.keys())
dict_keys(['topo'])

Who do I talk to?

If you encounter any problems or have questions, do not hesitate to raise an issue or make a pull request. Moreover, you can contact the devs directly:

• <bwinkel@mpifr.de>

Licenses

cysgp4 itself is published under GPL v3, an open-source license. The package is a Cython-powered wrapper of the sgp4lib library (by Daniel Warner) to compute satellite positions from two-line elements (TLE). The sgp4lib source code is licensed under Apache-2.0 license

The package is partly based on the Astropy-affiliated package template, which is under BSD 3-clause license.