Double Deep Reinforcement Learning Techniques for Low Dimensional Sensing Mapless Navigation of Terrestrial Mobile Robots.
Linda Dotto de MoraesVictor Augusto KichAlisson Henrique KollingJair Augusto BottegaRaul SteinmetzEmerson Cassiano da SilvaRicardo Bedin GrandoAnselmo Rafael CucklaDaniel Fernando Tello GamarraPublished in: CoRR (2023)
Keyphrases
- mobile robot
- low dimensional
- reinforcement learning
- obstacle avoidance
- indoor environments
- unstructured environments
- sensor fusion
- autonomous navigation
- unknown environments
- robot control
- real robot
- high dimensional
- navigation tasks
- outdoor environments
- path planning
- dimensionality reduction
- autonomous robots
- euclidean space
- potential field
- high dimensional data
- dynamic environments
- manifold learning
- data points
- input space
- collision free
- autonomous vehicles
- function approximation
- sensor networks
- multi robot
- motion control
- topological map
- collision avoidance
- robot navigation
- reinforcement learning algorithms
- principal component analysis
- feature space
- model free
- temporal difference
- multidimensional scaling
- state space
- mobile robotics
- linear dimensionality reduction
- robotic systems
- multi agent
- motion planning
- markov decision processes
- optimal policy
- sensory information
- multiple robots
- nonlinear dimensionality reduction
- transfer learning
- dynamic programming
- lidar data
- learning process
- image processing
- computer vision
- initially unknown
- learning algorithm