CASOG: Conservative Actor-Critic With SmOoth Gradient for Skill Learning in Robot-Assisted Intervention.
Hao LiXiao-Hu ZhouXiao-Liang XieShi-Qi LiuZhen-Qiu FengZeng-Guang HouPublished in: IEEE Trans. Ind. Electron. (2024)
Keyphrases
- robot assisted
- policy gradient
- skill learning
- gradient method
- minimally invasive
- robot arm
- reinforcement learning
- image guided
- action space
- function approximation
- optimal control
- single agent
- motion planning
- partially observable markov decision processes
- convergence rate
- state space
- image sequences
- control law
- dynamic environments
- decision makers