Robot Navigation in Constrained Pedestrian Environments using Reinforcement Learning.
Claudia Pérez-D'ArpinoCan LiuPatrick GoebelRoberto Martín-MartínSilvio SavaresePublished in: CoRR (2020)
Keyphrases
- robot navigation
- autonomous robots
- reinforcement learning
- continuous state
- autonomous mobile robot
- robot control
- mobile robot
- scene understanding
- initially unknown
- object detection
- pedestrian detection
- landmark recognition
- function approximation
- machine learning
- dynamic environments
- map building
- state space
- learning algorithm
- real time stereo
- model free
- real environment
- vision system
- robotic systems
- markov decision processes
- d scene
- partially observable
- real robot
- topological map