Self-supervised reinforcement learning for speaker localisation with the iCub humanoid robot.
Jonas Gonzalez-BillandonLukas GrasseMatthew S. TataAlessandra SciuttiFrancesco ReaPublished in: CoRR (2020)
Keyphrases
- humanoid robot
- reinforcement learning
- human robot interaction
- real robot
- motion planning
- multi modal
- biologically inspired
- imitation learning
- human robot
- function approximation
- manipulation tasks
- speech recognition
- model free
- fully autonomous
- motion capture
- walking speed
- cognitive architecture
- state space
- learning algorithm
- machine learning
- robot arm
- body movements
- motor control
- markov decision processes
- rough terrain
- temporal difference
- reinforcement learning algorithms
- reinforcement learning methods
- motor skills
- dynamic programming
- three dimensional