linhongbin / accel-challenge Goto Github PK

This is code of CUHK team for AccelNet2022, targeting to solve tasks of suturing autonomy in simulation environment. Please check out our project website

• Citation

@inproceedings{hlin2022Accel,
  title={{Open-source High-precision Autonomous Suturing Framework With Visual Guidance}},
  author={Lin, Hongbin and Li, Bin and Liu, Yunhui  and Kwok Wai Samuel Au},
  booktitle={IEEE Int. Conf. Robotics and Automation (ICRA) Workshop on ``A Panacea Or An Alchemy? Benefits And Risks Of Robot Learning In Medical Applications''},
  year={2022}
}

• open a terminal

  cd
  git clone https://github.com/linhongbin-ws/accel-challenge.git
  cd <path to accel-challenge>
  git clone https://github.com/collaborative-robotics/surgical_robotics_challenge.git

• Since our trained model is large, we put all trained model on the cloud. Please download them from google cloud and merge to model folder to the local directory <path to accel-challenge>/model/

• Install ambf

• create conda virtual environment python=3.7 .

  conda create -n accel_challenge python=3.7
  conda activate accel_challenge
  pip install torch==1.8.2 torchvision==0.9.2 torchaudio==0.8.2 --extra-index-url https://download.pytorch.org/whl/lts/1.8/cu111
  python -m pip install detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu111/torch1.8/index.html  
  pip install scikit-image
  pip install -r requirements.txt

• Install PyKDL on virtual environment from source, follow the instruction

  (note: make sure to uninstall the ros-kdl packages in the system before install PyKDL:

  sudo find / -iname PyKDL.so # this will print out all paths to PyKDL.so
  sudo rm -rf <path to>/PyKDL.so

• Install surgical_robot_challenge

  cd <path to surgical_robotics_challenge>/scripts/
  pip install -e .

• Install accel-challenge

  cd <path to accel-challenge>
  conda install -c conda-forge wxpython # deeplabcut depends on this package
  pip install -r requirements.txt
  pip install -e .

• modify a minor in original surgical_robotics_challenge, edit file <path to surgical_robotics_challenge>/scripts/surgical_robotics_challenge/launch_crtk_interface.py from

  import psm_arm
  import ecm_arm
  import scene

  to

  from surgical_robotics_challenge import psm_arm
  from surgical_robotics_challenge import ecm_arm
  from surgical_robotics_challenge import scene

• modify the file <path to accel-challenge>/accel_challenge/bash/user_var.sh, there are several path variables. You need to modify according to your enviroment path.

  AMBF_PATH="/home/ben/ssd/code/robot/ambf"
  SURGICAL_CHALLENGE_PATH='/home/ben/ssd/code/robot/accel-challenge/surgical_robotics_challenge'
  ANACONDA_PATH="/home/ben/anaconda3" 
  ENV_NAME="accel_challenge" # conda virtual environment name

• To run Challenge #1, please refer to README
• To run Challenge #2, please refer to README
• To run Challenge #3, please refer to README