Loading...

Optimal Control of a Quadcopter in Fast Descending Maneuvers Based on Reinforcement Learning

Azadi, Majid | 2022

143 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55525 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Fallah Rajabzadeh, Famida; Zohoor, Hassan; Nejat Pishkenari, Hossein
  7. Abstract:
  8. Quadrotors have limitation in performing fast descent maneuvers due to Vortex Ring State (VRS) region which make quadrotor unstable. In order to avoid entering VRS, a velocity constraint considered which it should be satisfied during this maneuver to guarantee a safe and stable fast descending maneuver by quadrotor. The purpose of this thesis is to overcome limitation in speed space of quadrotor in order to reduce the time of fast descending maneuvers by using Reinforcement Learning Techniques. A new cascade controller proposed which using PID in inner loop as a low level controller and DDPG as one of reinforcement learning techniques in outer loop as high level controller in order to generate an optimal time trajectory. The propose controller successfully design and implemented. The result shows time reduction of performing the maneuver up to 46.5% respect to normal descending without entering the VRS region. Also robustness of the controller has shown by applying different kind on disturbances and uncertainties. Finally the controller has implemented in flight simulation of Parrot Mambo project which results in 47.2% time reduction in performing fast descending maneuver without entering the VRS region
  9. Keywords:
  10. Quadcopter ; Quadrotor ; Deep Reinforcement Learning ; Optimal Control ; Deep Deterministic Policy Gradient ; Fast Descending Maneuver

 Digital Object List

 Bookmark

No TOC