A Physics-Informed Neural Networks Framework to Solve the Infinite-Horizon Optimal Control Problem.
Filippos FotiadisKyriakos G. VamvoudakisPublished in: CDC (2023)
Keyphrases
- optimal control
- infinite horizon
- neural network
- dynamic programming
- control problems
- finite horizon
- stochastic control
- optimal control problems
- control strategy
- long run
- markov decision processes
- average cost
- reinforcement learning
- partially observable
- markov decision process
- optimal policy
- single item
- probabilistic model
- multistage
- partially observable markov decision processes
- policy iteration