Learning continuous coupled multi-controller coefficients based on actor-critic algorithm for lower-limb exoskeleton.
Guangkui SongRui HuangHong ChengJing QuiQiming ChengShuai FanPublished in: Sci. China Inf. Sci. (2021)
Keyphrases
- actor critic
- learning algorithm
- policy gradient
- reinforcement learning
- gradient method
- optimal control
- approximate dynamic programming
- temporal difference
- np hard
- dynamic programming
- optimal solution
- cost function
- neuro fuzzy
- learning problems
- reinforcement learning algorithms
- average reward
- linear programming
- search space
- objective function
- function approximation
- policy iteration
- supervised learning