Introduce formula for Karney's inverse geodesic method #500
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This method is an integral part of Karney's solution to inverse geodesic problem. It is only invoked for points that lie on the meridian, or are close to it.
This fixes the inaccuracy caused during the calculation of Newton's starting point.
For details, please refer to Eq. (65) in, Geodesics on an ellipsoid of revolution, Charles F.F Karney, https://arxiv.org/abs/1102.1215
Conflicts (resolved): include/boost/geometry/util/math.hpp
…e structure The computed values from inverse method include distance, azimuth, reverse_azimuth, reduced_length, and geodesic_scale.
The compilation is successful with gcc version (7.2.0), but not with version (5.4.1). The accepted tolerance is set to (0.0000001). Currently, all tests are not passing, which indicates an error in the calculation. Additionally, some changes have been made in karney_inverse.hpp
These values are collected from GeodTest which is associated with GeographicLib: https://zenodo.org/record/32156 The conversion to C++ array format is done using this Python script: https://github.com/adl1995/boost-geometry-extra/blob/master/geographiclib-dataset-parse-inverse.py Geodesic scale (M12) is absent from the GeodTest dataset, so it is manually generated using GeographicLib using this C++ script: https://github.com/adl1995/boost-geometry-extra/blob/master/geographicLib-direct-antipodal.cpp
Contributor
Author
|
@vissarion, @awulkiew - I have resolved the bug which was causing test failures. Since all tests are passing now, please feel free to review the code. |
adl1995
changed the title
awulkiew
reviewed
Aug 13, 2018
awulkiew
reviewed
Aug 13, 2018
awulkiew
reviewed
Aug 13, 2018
awulkiew
reviewed
Aug 13, 2018
Member
|
@adl1995 are you willing to answer the above comments so we can merge this piece of code? I am ok with merging if the above comments are somehow addressed. |
Contributor
Author
|
@vissarion - Sorry for not being responsive on this PR. I will try and address the comments in the coming weeks so this can be merged. |
NaN was previously being used when the angle was greater than 90 deg. Since we now use the maximum angle value [-90, 90] this function is no longer required.
adl1995
force-pushed
the
feature/karney_inverse
branch
3 times, most recently
from
8bab57b to
05ae463
Compare
adl1995
force-pushed
the
feature/karney_inverse
branch
from
2bbf930 to
c5b684f
Compare
Member
|
Thanks Adeel for the long effort. NOTE: the operations are accurate (at least all the tests are passing) but I have noticed some performance issues, i.e. it is slower than We could merge now and resolve this issue in a different PR. @awulkiew @barendgehrels what do you think? |
Member
|
@adl1995 Thanks! @vissarion Yes, I'm ok with merging it now. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The implementation for Karney's inverse method is almost complete. The test cases are similar to Karney's direct method i.e. taken from GeographicLib. However, there a some bugs in the calculation which leads to failure in test cases.
Once those are resolved, you could review the code for any improvements.