https://github.com/um-arm-lab/focused-adaptation-dynamics

Code accompanying Focused Adaptation of Dynamics Models for Deformable Object Manipulation
https://github.com/um-arm-lab/focused-adaptation-dynamics

Last synced: 4 days ago
JSON representation

Code accompanying Focused Adaptation of Dynamics Models for Deformable Object Manipulation

• Host: GitHub
• URL: https://github.com/um-arm-lab/focused-adaptation-dynamics
• Owner: UM-ARM-Lab
• Created: 2023-03-13T19:30:55.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2023-03-14T14:04:54.000Z (over 1 year ago)
• Last Synced: 2023-08-24T16:14:16.585Z (about 1 year ago)
• Language: Jupyter Notebook
• Size: 358 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# FOCUS

This repository contains a snapshot of the code used for the paper "Focused Adaptation of Dynamcs Models for Deformable Object Manipulation".

The loss function shown in Equation (1) in the paper is implemented [here](https://github.com/UM-ARM-Lab/focused-adaptation-dynamics/blob/master/state_space_dynamics/src/state_space_dynamics/torch_udnn.py#L18). The hyperparameters used for experiments are in the folder `state_space_dynamics/hparams`. For example, the one used for simulated rope manipulation is [here](https://github.com/UM-ARM-Lab/focused-adaptation-dynamics/blob/master/state_space_dynamics/hparams/iterative_lowest_error_soft_online.hjson). The main scripts to run the full online learning method are in `link_bot_planning/scripts`:

- [online_rope_real.py](https://github.com/UM-ARM-Lab/focused-adaptation-dynamics/blob/master/link_bot_planning/scripts/online_rope_real.py)

- [online_rope_sim.py](https://github.com/UM-ARM-Lab/focused-adaptation-dynamics/blob/master/link_bot_planning/scripts/online_rope_sim.py)

- [online_water_sim.py](https://github.com/UM-ARM-Lab/focused-adaptation-dynamics/blob/master/link_bot_planning/scripts/online_water_sim.py)

For rope manipulation, we use Gazebo. For water simulation, we use PyFlex.

## Setup

The repository should be used as a reference impementation, as running the code requires a lot of setup, and extensive documentation is not provided. To start, clone this repo, as well as all the other UM-ARM-Lab repos you will need, such as `sdf_tools`, `arm_robots`, `arm_gazebo`, and possibly others.

```

# in the catkin workspace, inside the `src` folder

virtualenv --system-site-packages venv

source venv/bin/activate  # you will need to source the virtual environment whenever you want to run anything.

pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116

pip install -r requirements.txt

rosdep update

rosdep install -y -r --from-paths . --ignore-src

cd ..

catkin config -DOMPL_BUILD_PYBINDINGS=ON -DCATKIN_ENABLE_TESTING=OFF -DCMAKE_BUILD_TYPE=Release --merge-devel

catkin build

```