Low-level autonomous control and tracking of quadrotor using reinforcement learning - Proximal Policy Optimization

Instructions to run the code:

• Make sure the following libraries are installed:
  • Python 3.4x
  • Pytorch
  • Numpy
• go to the /code/ folder
• to train and run the hover task:
  • run $ python hover_train_and_run.py
• to train and run the trajectory following task:
  • run $ python traj_train_and_run.py
• The code can be run in collab if GPU environment to be used
  • Hover task
  • Trajectory following Task

Results:

Hovering Task:

Trajectory following Task: