IQL-TD-MPC: Implicit Q-Learning for Hierarchical Model Predictive Control.
Rohan ChitnisYingchen XuBobak HashemiLucas LehnertÜrün DoganZheqing ZhuOlivier DelalleauPublished in: ICRA (2024)
Keyphrases
- model predictive control
- mpc algorithm
- reinforcement learning
- reinforcement learning algorithms
- td learning
- control system
- predictive control
- temporal difference learning
- learning algorithm
- function approximation
- temporal difference
- specification language
- closed loop
- evaluation function
- state space
- eligibility traces
- model free
- control scheme
- relational calculus
- learning rate
- dynamic model
- complex objects
- optimal policy
- computational intelligence
- least squares
- knowledge base