SliceIt! - A Dual Simulator Framework for Learning Robot Food Slicing

A real2sim2real approach to learning robotic food slicing tasks with rigid robot manipulators.

Checkout:

• Paper
• Project page
• Video

Installation

Prerequisites

• Docker
• NVIDIA Container Toolkit

Install & Build

1. Download this repository and set up its submodules

git clone https://github.com/omron-sinicx/sliceit.git
cd sliceit
git submodule update --init --recursive

2. Build docker image

./BUILD-DOCKER-IMAGE.sh

3. Create and run docker container

./RUN-DOCKER-CONTAINER.sh

4. Build ROS environment inside container

o2ac-build-catkin-workspace

5. Install DiSECt as instructed in disect/README.md including downloading the dataset

pip install -r disect/requirements.txt
pip install -e /root/o2ac-ur/disect/meshing
pip install -e /root/o2ac-ur/disect
pip install -e /root/o2ac-ur/underlay_ws/src/ur_python_utilities/dependencies/tf2rl

Demo

Gazebo Only

• Visualization (simple env)

roslaunch o2ac_gazebo b_bot_gazebo.launch
rosrun ur3e_rl simple_env.py -e 1

Gazebo + DiSECt

• Visualization (simple env)

roslaunch o2ac_gazebo b_bot_gazebo.launch
cd ~/o2ac-ur/disect && python disect/cutting/ros_visualizer.py
rosrun ur3e_rl simple_env.py -e 3

• Policy learning

rosrun ur3e_rl tf2rl_sac.py -e 3

Authors:

• Cristian C. Beltran-Hernandez
• Nicolas Erbetti
• Masashi Hamaya

This repository is based on Team O2AC repository

@article{beltran2024sliceit,
  title={SliceIt!--A Dual Simulator Framework for Learning Robot Food Slicing},
  author={Beltran-Hernandez, Cristian C and Erbetti, Nicolas and Hamaya, Masashi},
  journal={arXiv preprint arXiv:2404.02569},
  year={2024}