Memory-based Deep Reinforcement Learning for Humanoid Locomotion under Noisy Scenarios.
Samuel ChenattiEsther Luna ColombiniPublished in: LARS/SBR/WRE (2022)
Keyphrases
- reinforcement learning
- degrees of freedom
- function approximation
- robot control
- state space
- markov decision processes
- real world
- noisy environments
- humanoid robot
- noisy data
- mobile robot
- optimal control
- action selection
- deep learning
- learning process
- learning capabilities
- autonomous learning
- multi agent reinforcement learning
- rough terrain
- noise free
- reinforcement learning algorithms
- motion planning
- optimal policy
- dynamic programming
- learning algorithm