AC4MPC: Actor-Critic Reinforcement Learning for Nonlinear Model Predictive Control.
Rudolf ReiterAndrea GhezziKatrin BaumgärtnerJasper HoffmannRobert D. McAllisterMoritz DiehlPublished in: CoRR (2024)
Keyphrases
- actor critic
- reinforcement learning
- mpc algorithm
- nonlinear model predictive control
- optimal control
- dynamic model
- temporal difference
- approximate dynamic programming
- reinforcement learning algorithms
- policy gradient
- neuro fuzzy
- model predictive control
- gradient method
- image based visual servoing
- function approximation
- state space
- policy iteration
- learning algorithm
- dynamic programming
- model free
- markov decision processes
- rl algorithms
- transfer learning
- policy gradient methods
- temporal difference learning
- machine learning
- optimal policy
- supervised learning
- state action
- control strategy
- evaluation function
- closed loop
- average reward
- control policy
- dynamical systems
- single agent