Adaptive Robotic Manipulator Simulation for Enhanced Feeding and Drinking Assistance

Implemented reinforcement learning algorithms (PPO, SAC and DDPG) across different manipulators simulating feeding and drinking tasks on OpenAI gym, which was published in IEEE INDICON-24 conference link of the paper

Getting Started

Dependencies

• Pytorch
• OpenAI gym
• CUDA
• Ubuntu 20.04

Installing

python3 -m pip install --upgrade pip
git clone https://github.com/s0um0r0y/spoon-feeder-ROS.git
cd spoon-feeder-ROS
python3 -m pip install -e .

Train a new control policy

• running training (PPO)

  python3 -m assistive_gym.learn --env "FeedingJaco-v1" --algo ppo --train --train-timesteps 20000 --save-dir ./trained_models_new/

• visualizing the trained model

  python3 -m assistive_gym.learn --env "FeedingJaco-v1" --algo ppo --render --seed 0 --load-policy-path ./trained_models_new/

Authors

Contributors names and contact info

• Soumo Roy email id
• Joel Viju email id
• Budhaditya Bhattacharyya email id

Video link

• youtube link
• google drive link

Citations

• Please consider citing us if you liked our project

@inproceedings{roy2024adaptive,
  title={Adaptive Robotic Manipulator Simulation for Enhanced Feeding and Drinking Assistance},
  author={Roy, Soumo and Viju, Joel and Bhattacharyya, Budhaditya},
  booktitle={2024 IEEE 21st India Council International Conference (INDICON)},
  pages={1--6},
  year={2024},
  organization={IEEE}
}

License

This project is licensed under the [MIT] License - see the LICENSE.md file for details

Acknowledgments

• I utilized assistive gym platform made by Dr. Zackory Erickson at CMU
• I would like to thank our professor Dr. Budhaditya Bhattacharya for his support, guidance and valuable time.
• This work was financialy supported by Vellore Institute of Technology (VIT), Vellore under the Faculty Seed Grant (RGEMS) (Sanction Order No.: SG20220001)