Skip to content

End-to-end optimal quadcopter control through Reinforcement Learning

Notifications You must be signed in to change notification settings

tudelft/optimal_quad_control_RL

Repository files navigation

optimal_quad_control_RL

Reinforcement learning for time optimal end-to-end quadcopter control

Main notebooks:

3D quad race:

  • gym environment of end-to-end quadcopter model
  • training pipeline
  • automatic c code generation

3D quad race INDI inner loop:

  • gym environment of quadcopter model with thrust and rate inputs
    • INDI controller is modeled as first order delay
  • training pipeline
  • automatic c code generation

About

End-to-end optimal quadcopter control through Reinforcement Learning

Resources

Stars

Watchers

Forks

Packages

No packages published