Policy-Iteration-Based Finite-Horizon Approximate Dynamic Programming for Continuous-Time Nonlinear Optimal Control.
Ziyu LinJingliang DuanShengbo Eben LiHaitong MaJie LiJianyu ChenBo ChengJun MaPublished in: IEEE Trans. Neural Networks Learn. Syst. (2023)
Keyphrases
- optimal control
- policy iteration
- approximate dynamic programming
- infinite horizon
- finite horizon
- average cost
- dynamic programming
- markov decision process
- markov decision processes
- single item
- reinforcement learning
- production planning
- optimal policy
- partially observable
- control strategy
- brownian motion
- control policies
- actor critic
- average reward
- markov decision problems
- inventory level
- search algorithm
- learning algorithm