PPO-based Flapping Wing MAV Controller

Reinforcement Learning for Robust Locomotion Control of Flapping Wing MAV

Description

• Hard Coded Version: MujocoEnv, Gymnasium, and Stable-Baselines3 are used
• Mujoco built-in fluid dynamics is used

Getting Started

Dependencies

• gymnasium==0.29.1
• matplotlib==3.8.2
• mpi4py==3.1.5
• numba==0.58.1
• numpy==1.26.3
• scipy==1.12.0
• stable_baselines3==2.2.1
• tensorflow==2.15.0
• tensorflow_macos==2.15.0

Executing program

• Execute train/run.py
• Logs and models will be saved at train/logs and train/saved_models respectively

cd train
python run.py

Authors

Mintae Kim
Jiaze Cai
Jack Leckert

Version History

• 0.1.0
  • Initial Release

License

This project is licensed under the MIT License - see the LICENSE file for details