VesNet-RL: Simulation-Based Reinforcement Learning for Real-World US Probe Navigation.
Yuan BiZhongliang JiangYuan GaoThomas WendlerAngelos KarlasNassir NavabPublished in: IEEE Robotics Autom. Lett. (2022)
Keyphrases
- reinforcement learning
- real world
- function approximation
- model free
- reinforcement learning algorithms
- optimal policy
- wide range
- direct policy search
- markov decision processes
- multi agent
- control problems
- data mining
- state space
- learning algorithm
- rl algorithms
- real robot
- synthetic data
- autonomous learning
- reinforcement learning methods
- temporal difference
- partially observable
- action selection
- robot navigation
- machine learning
- approximate dynamic programming
- supervised learning
- data sets
- reward function
- partially observable domains
- state and action spaces
- policy search
- continuous state
- temporal difference learning
- learning agents
- policy iteration
- learning capabilities