Model-free H∞ tracking control for de-oiling hydrocyclone systems via off-policy reinforcement learning.

Shaobao Li Petar Durdevic Zhenyu Yang

Published in: Autom. (2021)

Keyphrases

model free
reinforcement learning
reinforcement learning algorithms
tracking control
function approximation
temporal difference
state space
policy iteration
learning algorithm
neural network
machine learning
expert systems
markov decision processes
closed loop
dynamic model
policy evaluation