Skip to content

Ericcsr/DiffSRL

BranchesTags
 
 

Repository files navigation

DiffSRL: Learning Dynamic-Aware State representation for Control via Differentiable Simulation

Pre-released code bug expected.

Install

  • Run cd ChamferDistancePytorch
  • Install python3 -m pip install -e .
  • Run cd ..
  • Install python3 -m pip install -e .

Enjoy the pretrained model

Model Free Reinforcement Learning on Chopsticks or Rope

  • Run python3 -m plb.algorithms.solve --algo td3 --env_name [Chopsticks-v1/Rope-v1] --exp_name enjoy --model_name rope/encoder --render

Model Based Policy Optimization on Chopsticks or Rope

  • Run python3 -m plb.algorithms.solve --algo torch_nn --env_name [Chopsticks-v1/Rope-v1] --exp_name enjoy --model_name rope/encoder --render

Training new model

Collect data from new environment

  • Run python3 -m plb.algorithms.solve --algo collect --env_name [EnvName-version] --exp_name [new_environment] Which will collect raw data and stored in raw_data folder.
  • Run python3 preprocess.py --dir raw_data/[new_environment]to pre-process data and the preprocessed npz file will be stored in data with the name of [new_environment].

Running State Representation learning using new dataset

  • Run python3 plb.algorithms.solve --env_name [EnvName-version] --exp_name [EnvName-version] --exp_name learn_latent --lr 1e-5 The encoder weight will be saved in pretrained_model

Experiment result

All experiment result are rendered from policy trained with MBPO

  • Picking up a rope
Image
  • Wrapping a rope around a cylinder
Image

Simulation Transfer to Real Experiments

Rope Experiment (The coordinate frame has been flipped to avoid occlusion)

Image
Image
#### Chopsticks Experiment
Image
Image

About

Official Project Webpage for paper "DiffSRL: Learning Dynamic-aware State Representation for Control via Differentiable Simulation"

Resources

Readme
Activity

Stars

12 stars

Watchers

2 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 90.7%
  • Cuda 3.3%
  • Jupyter Notebook 2.7%
  • Shell 1.8%
  • C++ 1.5%