Policy gradient based Reinforcement Learning for real autonomous underwater cable tracking.
Andres El-FakdiMarc CarrerasPublished in: IROS (2008)
Keyphrases
- reinforcement learning
- optimal policy
- autonomous learning
- policy search
- real time
- markov decision process
- target tracking
- model free
- visual tracking
- particle filter
- reinforcement learning problems
- state space
- action selection
- policy iteration
- function approximators
- reinforcement learning algorithms
- state and action spaces
- fiber optic
- markov decision problems
- rl algorithms
- action space
- partially observable environments
- real robot
- temporal difference
- infinite horizon
- particle filtering
- robotic systems
- appearance model
- kalman filter
- markov decision processes
- object tracking
- learning algorithm