-
Notifications
You must be signed in to change notification settings - Fork 38
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Restructure ML component #771
Conversation
…put in ML component
- define the MANNFootState class - move all the quantities required to reset the state in MANNAutoregressive::AutoregressiveState - simplify the generation of the initial AutoregressiveState required to start the trajectory generation
- eliminate the need to pass foot positions in MANNTrajectoryGenerator::setInitialState - enhance MANNTrajectoryGenerator with reset capabilities similar to MANNAutoregressive - correct the position and time scaling of MANNTrajectoryGenerator to accurately scale the CoM, base, and feet positions - update the Python bindings
a01ed4b
to
46711c4
Compare
46711c4
to
6bfde64
Compare
6bfde64
to
8287ac6
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Some preliminary comments
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reading just the header, I was a bit puzzled by the word dummy
. I interpret dummy
usually as a placeholder, so it is not clear to me why you want to generate a dummy
? Can dummy
be removed is the resulting output is actually well formed?
Reading the |
Co-authored-by: Stefano Dafarra <stefano.dafarra@gmail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Mainly fixed typos and added suggestions for the text of the two examples!
…and MANNAutoregressive examples
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Stefano Dafarra <stefano.dafarra@gmail.com>
Co-authored-by: Stefano Dafarra <stefano.dafarra@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
In ae62976 I improved the documentation regarding the dummy |
Hi @paolo-viceconte and @S-Dafarra I removed the hardcoded size of the vectors in MANN e19b84e |
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Co-authored-by: Paolo Viceconte <paolomariaviceconte@gmail.com>
Hi @S-Dafarra and @paolo-viceconte if you are fine with the changes we can merge it |
This PR is the outcome of the refactoring carried out with @paolo-viceconte
In detail, we implemented
MANN::generateDummyMANNOutput
andMANN::generateDummyMANNInput
in theML
component. Additionally, we restructuredMANNAutoregressive
for improved reset handling by introducing theMANNFootState
class and consolidating reset-related quantities inMANNAutoregressive::AutoregressiveState
. This simplifies the generation of the initialAutoregressiveState
required to start trajectory generation. Concurrently, we restructuredMANNTrajectoryGenerator
by eliminating the need to pass foot positions in setInitialState. We enhancedMANNTrajectoryGenerator
with reset capabilities similar toMANNAutoregressive
and corrected the position and time scaling to ensure accurate scaling of the CoM, base, and feet positions. The Python bindings have been updated accordingly.Last but not least I added two examples that show how to use
MANNAutoregressive
andMANNTrajectoryGenerator
MANNAutoregressive
Screencast.from.27-11-2023.18.16.54.webm
MANNTrajectoryGenerator
Screencast.from.27-11-2023.18.19.00.webm