Data-driven virtual reference feedback tuning and reinforcement Q-learning for model-free position control of an aerodynamic system.
Mircea-Bogdan RadacRadu-Emil PrecupRaul-Cristian RomanPublished in: MED (2016)
Keyphrases
- model free
- reinforcement learning
- impedance control
- position control
- reinforcement learning algorithms
- function approximation
- feedback loop
- control strategies
- dynamic model
- control scheme
- robotic manipulator
- learning algorithm
- policy iteration
- temporal difference
- state space
- force control
- visual servoing
- robot arm
- virtual reality
- rl algorithms
- closed loop
- optimal policy
- control system
- action selection
- dynamic programming
- real time
- linear combination
- machine learning
- visual feedback
- optimal control
- degrees of freedom
- markov decision processes
- supervised learning
- neural network