A variable-barrier-function-based prescribed finite-time bounded-H∞ reinforcement learning optimal tracking control strategy without dependence on initial condition.
Ying LiuXiaohua LiSongnan ZouHui LiuYang LiuPublished in: J. Frankl. Inst. (2024)
Keyphrases
- control strategy
- optimal control
- reinforcement learning
- control strategies
- control algorithm
- control method
- operating conditions
- control system
- mathematical model
- control scheme
- control policy
- control law
- continuous functions
- dynamic programming
- fuzzy logic control
- speed control
- force control
- predictive control
- real time
- biped robot
- function approximators
- pi controller
- fuzzy controller
- sliding mode control
- controller design
- function approximation
- state and action spaces
- multi agent
- lotka volterra