Path Following Optimization for an Underactuated USV Using Smoothly-Convergent Deep Reinforcement Learning.
Yujiao ZhaoXin QiYong MaZhixiong LiReza MalekianMiguel Ángel SoteloPublished in: IEEE Trans. Intell. Transp. Syst. (2021)
Keyphrases
- reinforcement learning
- function approximation
- optimization problems
- optimization algorithm
- global optimization
- learning algorithm
- motion planning
- constrained optimization
- optimization methods
- supervised learning
- optimization process
- model free
- temporal difference
- discrete optimization
- learning problems
- markov decision processes
- optimal policy
- deep learning
- mechanical systems
- policy search