Retro-RL: Reinforcing Nominal Controller With Deep Reinforcement Learning for Tilting-Rotor Drones.
I. Made Aswin NahrendraChristian TirtawardhanaByeongho YuEungchang Mason LeeHyun MyungPublished in: CoRR (2022)
Keyphrases
- long run
- control policy
- reinforcement learning
- optimal policy
- actor critic
- operating conditions
- optimal control
- initial state
- state space
- mathematical model
- reinforcement learning algorithms
- adaptive control
- fault diagnosis
- markov decision processes
- induction motor
- closed loop
- model free
- control system
- rl algorithms
- control strategies
- function approximation
- control algorithm
- transfer learning
- multi agent
- policy search
- control method
- control strategy
- dynamic programming
- traffic signal
- learning process
- markov decision process
- learning algorithm
- temporal difference learning
- direct policy search
- control scheme
- reinforcement learning methods
- action selection
- pid controller
- neural network
- machine learning