Matthias Grob
|
84cd7483ae
|
Quaternion: added constructor which generates the shortest rotation that maps one vector to another
including tedious corner case handling for parallel vectors with 180 degree rotations
|
7 years ago |
|
Matthias Grob
|
5130da206a
|
Vector: added norm_squared() because sometimes you can safe the sqrt operation
|
7 years ago |
|
Matthias Grob
|
41ad2bdea5
|
Matrix: added copyToRaw method to allow copying to a pointer
because most uORB messages still contain all components of a vector one by one after each other
|
7 years ago |
|
Matthias Grob
|
1dffd5930b
|
Matrix: add copyTo copying data to an array and copyToColumnMajor which does the same but with column-major order
same functionality explicitly for quaternions can be deleted
|
7 years ago |
|
Matthias Grob
|
9e59691e43
|
Vector: Additional normalization with check for zero norm because it occurs so many times in applications
|
7 years ago |
|
Matthias Grob
|
f4243160e2
|
Quaternion: changed comments because of typos and unclear inconsistent indexing
|
7 years ago |
|
Matthias Grob
|
ee2219b836
|
Quaternion: replace conversion in rotate() with AxisAngle call
|
7 years ago |
|
Matthias Grob
|
b241cf5c95
|
Quaternion: Adjusted rotate() to the Hamilton convention (which we switched to)
Note: This error was not caught by a test because the test included only trivial cases which do not explore non-commuting quaternions.
|
7 years ago |
|
Matthias Grob
|
6b1fea76d0
|
Quaternion: added and adjusted comments to further explain conjugation and derivative functionality
|
7 years ago |
|
Matthias Grob
|
0527471a52
|
Matrix: adjust buffer size calculation to account for additional characters
Note: If the buffer is too small there's no memory corruption because of the snprintf limit but part of the output gets missing.
|
7 years ago |
|
Matthias Grob
|
0a772f59dd
|
Quaternion: added direct efficient body z-axis calculation with test
|
8 years ago |
|
Matthias Grob
|
976461eb0f
|
Dcm: more efficient conversion from quaternion, extend reuse of multiplications
|
8 years ago |
|
Matthias Grob
|
baf54ad29f
|
quaternion: correcting comments
^(-1) went missing and some phrase was inprecise
|
8 years ago |
James Goppert
|
e595ebb9a7
|
Switch to Hamilton quaternions and add Cholesky decomposition.
|
8 years ago |
Nate Weibley
|
471e96ff6f
|
Provide const dataptr access
|
8 years ago |
|
Nate Weibley
|
2ad3ec46b9
|
Mark simple min, max, abs methods as const
|
8 years ago |
|
Nate Weibley
|
66e1b406b8
|
Remove artifical need to virtualize dtors
|
8 years ago |
Pavel Kirienko
|
499b897e5f
|
Style fix
|
8 years ago |
|
Pavel Kirienko
|
99b44c0242
|
NuttX math lib workaround
|
8 years ago |
|
Pavel Kirienko
|
8dbe4a7531
|
NuttX workaround
|
8 years ago |
|
Pavel Kirienko
|
23def31d21
|
Fixed stdlib imports
|
8 years ago |
|
Pavel Kirienko
|
9ebf5f89db
|
Removed all use of C library from the matrix namespace
|
8 years ago |
|
Pavel Kirienko
|
552dad40a1
|
Fixed inclusions in matrix/
|
8 years ago |
Daniel Agar
|
cfa68c2196
|
clang-tidy trivial cleanup
|
8 years ago |
|
Daniel Agar
|
f00edc9442
|
update to astyle 2.06 and fix formatting
|
8 years ago |
|
Daniel Agar
|
68c7cc5bfd
|
Quaternion add copyTo
|
8 years ago |
|
Daniel Agar
|
e2211c5867
|
QuRT use __builtin_isfinite
|
8 years ago |
Lorenz Meier
|
2283e6946a
|
Matrix inversion: Ensure that null check is done against the same type
|
8 years ago |
|
James Goppert
|
843be9418b
|
Rename isfinite to is_finite to avoid name conflicts.
|
8 years ago |
|
James Goppert
|
63aea23f9e
|
Add cholesky decomp, Closes #30, and dynamic print buf
|
8 years ago |
|
James Goppert
|
a154e14439
|
Fix coverage for inverse.
|
8 years ago |
|
James Goppert
|
db6dfeafbe
|
Add helper_functions include for inverse and in general header
|
8 years ago |
|
James Goppert
|
230e84702a
|
Fix unit test, create matrix isfinite.
|
8 years ago |
Siddharth Bharat Purohit
|
7e3eff7b2d
|
remove unnecessary duplicate matrices from inverse
|
8 years ago |
|
Matthias Grob
|
2f4427a923
|
Quaternion: Implemented the largest pivot element method for conversion from Dcm to Quaternion to avoid numerical problems
|
8 years ago |
|
Matthias Grob
|
bb1a321639
|
Changed Dcm to inherit from SquareMatrix because it is square and we can use the trace method
|
8 years ago |
|
James Goppert
|
7eff04742c
|
Fix for zero check on quat from dcm ctor.
|
8 years ago |
|
James Goppert
|
74a120f554
|
Handle some divide by zero edge cases on init.
|
8 years ago |
|
James Goppert
|
945edbc58c
|
Ensure direction is always defined for axis angle.
|
8 years ago |
|
James Goppert
|
8a94f0f8ac
|
Fix axis angle fabsf usage.
|
8 years ago |
|
James Goppert
|
f3e478cbff
|
Fix matrix inverse pivotting logic.
|
8 years ago |
James Goppert
|
34fccdd680
|
add renormalize method (#32)
|
8 years ago |
|
James Goppert
|
3aeb099249
|
Revert "Use Hamilton quaternion definition."
This reverts commit 19554c4470.
|
9 years ago |
|
James Goppert
|
0eb8aa0c0b
|
Support more of mathlib function for easier swap in px4.
|
9 years ago |
|
James Goppert
|
558777f34c
|
Fix formatting.
|
9 years ago |
|
James Goppert
|
262a715d90
|
Fix some when building in px4.
|
9 years ago |
|
James Goppert
|
6197515a51
|
Fix shadow warnings.
|
9 years ago |
|
James Goppert
|
4f13809420
|
Added axis angle accessors, removed == operator.
|
9 years ago |
|
James Goppert
|
19554c4470
|
Use Hamilton quaternion definition.
|
9 years ago |
|
James Goppert
|
e6a6b4680c
|
Use quaternion multiplaction for quaternion derivative.
|
9 years ago |
