Deep Reinforcement Learning based Robot Navigation in Dynamic Environments using Occupancy Values of Motion Primitives.
Neset Ünver AkmandorHongyu LiGary LvovEric DuselTaskin PadirPublished in: IROS (2022)
Keyphrases
- dynamic environments
- robot navigation
- reinforcement learning
- autonomous robots
- mobile robot
- reinforcement learning agents
- autonomous agents
- path planning
- changing environment
- scene understanding
- motion primitives
- single agent
- state space
- topological map
- dynamic programming
- simultaneous localization and mapping
- autonomous navigation
- real environment
- machine learning
- spatio temporal
- human motion
- markov decision processes
- optimal policy