Constrained reinforcement learning with statewise projection: a control barrier function approach.
Xinze JinKuo LiQing-Shan JiaPublished in: Sci. China Inf. Sci. (2024)
Keyphrases
- reinforcement learning
- control policy
- control problems
- optimal control
- robot control
- learning algorithm
- function approximation
- real time
- robotic control
- control system
- function approximators
- control strategies
- machine learning
- temporal difference
- adaptive control
- control strategy
- piecewise linear
- control method
- action selection
- evaluation function
- data acquisition
- state space
- dynamic programming
- learning process
- state action
- policy gradient
- transition model
- multi agent