AUV Pipeline Following using Reinforcement Learning.
Sigurd Aksnes FjerdingenErik KyrkjebøAksel Andreas TransethPublished in: ISR/ROBOTIK (2010)
Keyphrases
- reinforcement learning
- function approximation
- autonomous underwater vehicle
- state space
- pipeline architecture
- temporal difference
- model free
- multi agent
- markov decision processes
- optimal policy
- machine learning
- temporal difference learning
- control problems
- reinforcement learning algorithms
- real world
- dynamic environments
- markov chain
- supervised learning
- dynamic programming
- learning process
- reward function
- data streams
- cooperative
- information systems
- processing pipeline
- robotic control
- learning algorithm