Dynamic Interval Restrictions on Action Spaces in Deep Reinforcement Learning for Obstacle Avoidance.
Tim GramsPublished in: CoRR (2023)
Keyphrases
- action space
- obstacle avoidance
- reinforcement learning
- mobile robot
- continuous state
- path planning
- markov decision processes
- state and action spaces
- state space
- visually guided
- real valued
- markov decision problems
- dynamic environments
- motion planning
- action selection
- neural network
- reinforcement learning algorithms
- stochastic processes
- autonomous vehicles
- state action
- bayesian networks
- function approximation
- function approximators
- dynamic programming
- evolutionary algorithm
- continuous state spaces
- search algorithm
- skill learning