Fix getAxis implementation in Revolute and Prismatic Joint and documentation cleanup #339
Conversation
|
I just noticed that I also fixed the computation of the motion subspace vectors for the prismatic joint in |
|
The test that revealed this would be included in the following PR, addressing #340 . |
|
I need to proceed with #340 , so I am going to merge this. But @DanielePucci please remember to get someone to understand this PR! |
|
Yes @traversaro : @nunoguedelha, @fjandrad , and @diegoferigo can you make a small team to dig into this pull request? Thanks! |
|
The idea of the small team is to speed up the comprehension of the PR, which shall not take more than one-working day, and thus not requiring any blocking of the on-going activities. This micro-activity has low priority, but it must be accomplished by mid of September. Thank you guys. |
|
In view of the ICRA deadline, let's do it by the end of September. |
traversaro
force-pushed
the
fixGetAxis
branch
2 times, most recently
from
fc6862e
to
7b61e16
Compare
…t and documentation cleanup The direction of the axis in a joint depends on which body you consider parent and which body you consider child . See "Minimal formulation of joint motion for biomechanisms", Section 2.4 and Traversaro's PhD Thesis, Chapter 3. This is a small change, but it is extremly important. I hope that the additional documentation will help document it.
|
@nunoguedelha @diegoferigo @traversaro Shall we try the small team PR review? |
The direction of the axis in a joint depends on which body
you consider parent and which body you consider child .
Seth, A., Sherman, M., Eastman, P., & Delp, S. (2010).
Minimal formulation of joint motion for biomechanisms.
Nonlinear Dynamics, 62(1), 291-303.
https://nmbl.stanford.edu/publications/pdf/Seth2010.pdf
Section 2.4
and
"Modelling, Estimation and Identification of Humanoid Robots Dynamics"
Traversaro - Section 3.2
https://traversaro.github.io/preprints/traversaro-phd-thesis.pdf
for more details.
This is a small change, but it is extremely important.
I hope that the additional documentation will help document it.
A few additional methods for the Direction and the Axis class leaked in from the bigger PR I am preparing to address #340 , I hope they do not complicate the review too much.