IQL-TD-MPC: Implicit Q-Learning for Hierarchical Model Predictive Control.
Rohan ChitnisYingchen XuBobak HashemiLucas LehnertUrun DoganZheqing ZhuOlivier DelalleauPublished in: CoRR (2023)
Keyphrases
- model predictive control
- mpc algorithm
- reinforcement learning
- reinforcement learning algorithms
- td learning
- predictive control
- control system
- function approximation
- temporal difference learning
- temporal difference
- learning algorithm
- specification language
- state space
- eligibility traces
- relational calculus
- model free
- markov decision processes
- closed loop
- database
- dynamic model
- action selection
- control scheme
- evaluation function
- optimal policy
- least squares
- query language
- dynamic programming
- decision making
- real time