Symmetric actor-critic deep reinforcement learning for cascade quadrotor flight control.
Haoran HanJian ChengZhilong XiMaolong LvPublished in: Neurocomputing (2023)
Keyphrases
- actor critic
- reinforcement learning
- flight control
- optimal control
- temporal difference
- control law
- reinforcement learning algorithms
- approximate dynamic programming
- policy gradient
- gradient method
- function approximation
- adaptive control
- neuro fuzzy
- policy iteration
- markov decision processes
- state space
- average reward
- model free
- dynamic programming
- unmanned aerial vehicles
- dynamical systems
- machine learning
- learning problems
- reinforcement learning methods
- multi agent
- action selection
- reward function
- learning tasks
- linear program
- rl algorithms
- optimal policy
- expert systems
- policy gradient methods