Learning to Fly - a Gym Environment with PyBullet Physics for Reinforcement Learning of Multi-agent Quadcopter Control.
Jacopo PaneratiHehui ZhengSiqi ZhouJames XuAmanda ProrokAngela P. SchoelligPublished in: IROS (2021)
Keyphrases
- reinforcement learning
- multi agent
- learning algorithm
- learning process
- learning agents
- control problems
- learning problems
- robot control
- autonomous robots
- function approximation
- supervised learning
- multi agent environments
- real time
- learning tasks
- learning systems
- online learning
- active learning
- multi agent systems
- cooperative
- learning community
- bayesian networks
- complex environments
- artificial intelligence
- actor critic
- agent behavior
- reinforcement learning agents
- active exploration
- learning capabilities
- reinforcement learning algorithms
- temporal difference
- transfer learning
- dynamic environments
- machine learning
- neural network