Performance-Driven Controller Tuning via Derivative-Free Reinforcement Learning.
Yuheng LeiJianyu ChenShengbo Eben LiSifa ZhengPublished in: CDC (2022)
Keyphrases
- derivative free
- reinforcement learning
- neural learning
- gradient method
- unconstrained optimization
- control system
- optimization methods
- pid controller
- fan beam
- objective function
- convergence rate
- control strategy
- closed loop
- step size
- control scheme
- optimization problems
- nonlinear optimization
- supervised learning
- dynamic programming