Identification and control using a hybrid reinforcement learning system.
Peter M. MillsMoses O. TadéAlbert Y. ZomayaPublished in: Int. J. Comput. Simul. (1995)
Keyphrases
- reinforcement learning
- control problems
- robot control
- control strategy
- optimal control
- receding horizon
- function approximation
- control system
- control strategies
- mathematical model
- control policy
- model free
- action selection
- process control
- artificial neural networks
- data sets
- hybrid approaches
- feedback control
- learning algorithm
- temporal difference
- database
- least squares
- learning problems
- learning tasks