Robust Parameter Estimation and Tracking through Lyapunov-based Actor-Critic Reinforcement Learning.
Thomas RudolfJoshua RansiekStefan SchwabSören HohmannPublished in: IECON (2022)
Keyphrases
- actor critic
- reinforcement learning
- robust parameter estimation
- policy gradient
- temporal difference
- optimal control
- adaptive control
- approximate dynamic programming
- reinforcement learning algorithms
- tracking error
- neuro fuzzy
- function approximation
- dynamical systems
- particle filter
- lyapunov function
- state space
- rl algorithms
- gradient method
- policy iteration
- nonlinear systems
- control law
- control scheme
- markov decision processes
- average reward
- stability analysis
- supervised learning
- multi agent
- sufficient conditions
- closed loop
- kalman filter
- reinforcement learning methods
- temporal difference learning
- learning algorithm
- monte carlo
- linear program
- policy gradient methods