google-deepmind / google-deepmind/dm_control
Advice modeling simple block sliding on floor
- Langage dominant
- Python
- Étoiles
- 4.7k
- Forks
- 764
- Métriques de merge des PR
- Aucune PR mergée en 30 j
Description
I'm getting started with dm_control and am trying to model a bare bones environment: a box sliding around a floor.
To do this, I adapted from [mujoco-worldgen](https://github.com/openai/mujoco-worldgen) just a box with 3 slide joints and corresponding actuators. I also use the floor arena and then edited down locomotion walker/target tasks to be as simple as possible.
XML here. Note that if I don't include joint armature = 1, then it immediately fails w/ `Nan, Inf or huge value in QACC at DOF 5` at time t=0.000. When I include it though, the box does not move.
```
```
Full zip of code is attached.
[ice 2.zip](https://github.com/deepmind/dm_control/files/9504813/ice.2.zip)
Guide de contribution
Ouvrir le guide de contribution
Piste de recherche
Start by reproducing the XML model and attached ice 2.zip, focusing on the MuJoCo simulation at time 0.000 and the QACC error. Read the dm_control task and arena setup used in the archive before deciding whether this is a code defect or a modeling issue. Done means identifying a reproducible cause and documenting or implementing a validated fix for the box's movement.
Rédigé par le modèle d'indexation à partir du texte de l'issue.
Évaluation
- Stack technique
- python
- Domaine
- machine-learning
- Type d'issue
- Bug
- Difficulté
- 5/5
- Temps estimé
- Plus d'une semaine
- Activité
- À l'abandon
- Clarté
- À clarifier
- Accessibilité débutants
- 20/100