Navigation robot training with Deep Q-Learning monitored by Digital Twin.
Madson Rodrigues LemosValtemar Fernandes CardosoMario OtaniRivelino Da Costa NunesVandermi João da SilvaVicente Ferreira de Lucena JuniorPublished in: ICCE (2022)
Keyphrases
- potential field
- obstacle avoidance
- mobile robot
- robot navigation
- path planning
- navigation tasks
- autonomous navigation
- multi robot
- reinforcement learning
- visual navigation
- service robots
- multi agent
- cooperative
- machine learning
- unknown environments
- learning algorithm
- real robot
- indoor environments
- vision system
- robot soccer
- human robot interaction
- rough terrain
- training process
- biologically inspired
- function approximation
- real time
- topological map
- outdoor environments
- robot manipulators
- humanoid robot
- motor learning
- training examples
- motor skills
- deep architectures
- reinforcement learning algorithms
- autonomous robots
- motion planning
- supervised learning
- state space
- digital libraries