Extending the OpenAI Gym for robotics: a toolkit for reinforcement learning using ROS and Gazebo.
Iker ZamoraNestor Gonzalez LopezVictor Mayoral VilchesAlejandro Hernández CorderoPublished in: CoRR (2016)
Keyphrases
- reinforcement learning
- perception action
- robot control
- function approximation
- artificial intelligence
- computer vision
- real robot
- learning process
- temporal difference learning
- temporal difference
- action selection
- model free
- state space
- continuous state
- reinforcement learning algorithms
- stochastic approximation
- markov decision processes
- relational reinforcement learning
- robotic systems
- imitation learning
- learning problems
- transfer learning
- robotic control
- supervised learning
- autonomous navigation
- learning algorithm
- learning agents
- educational robotics
- multi agent reinforcement learning
- cognitive robotics
- neural network
- function approximators
- mobile robotics
- control problems
- partially observable
- reward function
- machine intelligence
- optimal policy
- vision system
- dynamic programming
- learning environment
- information retrieval