Development and Validation of Active Roll Control based on Actor-critic Neural Network Reinforcement Learning.
Matthias BahrSebastian ReichertsPhilipp Maximilian SiebergLuca MorssDieter SchrammPublished in: SIMULTECH (2019)
Keyphrases
- actor critic
- reinforcement learning
- neural network
- optimal control
- control problems
- policy gradient
- function approximation
- temporal difference
- reinforcement learning algorithms
- neuro fuzzy
- state space
- approximate dynamic programming
- control policy
- gradient method
- control system
- adaptive control
- learning algorithm
- artificial neural networks
- average reward
- policy iteration
- control strategies
- model free
- control method
- control strategy
- radial basis function
- markov decision processes
- action selection
- dynamic programming
- mobile robot
- dynamical systems
- back propagation
- temporal difference learning
- dynamic environments
- supervised learning
- fuzzy logic