⚠️ Deprecation Notice

This repository is deprecated and is no longer actively maintained.

The repository will be publicly archived, meaning it will remain accessible in a read-only state. Please consider using SofaGym and LapGym as they have much better implementations and are actively maintained.

---

Soft Tissue Manipulation

This repo contains the main code for the paper Sim-to-Real Surgical Robot Learning and Autonomous Planning for Internal Tissue Points Manipulation Using Reinforcement Learning.

Requirements

• SOFA Framework==v21.12.00 with SofaPython3 plugin
• stable-baselines3==1.6.1
• gym==0.21.0

Usage

Please refer to test_env.py.

env = SkinTissueEnvContinuousRandom(obs_sequence_length=1, params=params, render_mode="pyplot", randomize=True)
env.reset()

for _ in range(20):
    act = env.action_space.sample()
    env.step(act)

Citation

@article{ou2023sim,
  title={Sim-to-Real Surgical Robot Learning and Autonomous Planning for Internal Tissue Points Manipulation Using Reinforcement Learning},
  author={Ou, Yafei and Tavakoli, Mahdi},
  journal={IEEE Robotics and Automation Letters},
  volume={8},
  number={5},
  pages={2502--2509},
  year={2023},
  publisher={IEEE}
}