Learn to efficiently exploit cost maps by combining RRT* with Reinforcement Learning.
Riccardo FranceschiniMatteo FumagalliJulián Cayero BecerraPublished in: SSRR (2022)
Keyphrases
- reinforcement learning
- high cost
- storage cost
- machine learning
- learning algorithm
- learning agent
- complex domains
- function approximation
- function approximators
- optimal policy
- total cost
- temporal difference learning
- markov decision processes
- supervised learning
- dynamic programming
- case study
- communication cost
- neural network
- minimum cost
- combining multiple
- model free
- cost sensitive
- temporal difference
- genetic algorithm
- cost reduction
- internal memory