Koopman-based Policy Iteration for Robust Optimal Control.
Alexander KrolickiSarang SutavaniUmesh VaidyaPublished in: ACC (2022)
Keyphrases
- optimal control
- policy iteration
- infinite horizon
- dynamic programming
- control problems
- markov decision processes
- control strategy
- reinforcement learning
- average cost
- average reward
- optimal control problems
- actor critic
- least squares
- policy evaluation
- fixed point
- optimal policy
- approximate dynamic programming
- linear programming
- markov decision process
- multistage
- supervised learning
- real time
- finite horizon