Skip to content
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

cost too much time #4

Open
dujinyu opened this issue Jun 11, 2020 · 1 comment
Open

cost too much time #4

dujinyu opened this issue Jun 11, 2020 · 1 comment

Comments

@dujinyu
Copy link

dujinyu commented Jun 11, 2020

hello!
I met some problems when i ran your code.

first problem:
too much time is needed, the code did not converge after running half an hour.

second problem:
you said that you use ilqr, but i still find the Second derivative of dynamics. In ilqr_gym.py, self._f_xx, self._f_uu, self._f_ux exist. This is inconsistent with the ilqr method omitting the second derivative of dynamics.

third problem:
it seems that the animation of gym does not work. the pole does not move continuously.
image

@neka-nat
Copy link
Owner

Hi @dujinyu ,

Thank you for your feedback.
The slow computation and drawing discontinuity is due to the computational load of autograd.
As for the algorithm, it will be renamed to DDP instead of iLQR.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants