Data-Driven Human-Robot Interaction Without Velocity Measurement Using Off-Policy Reinforcement Learning.
Yongliang YangZihao DingRui WangHamidreza ModaresDonald C. WunschPublished in: IEEE CAA J. Autom. Sinica (2022)
Keyphrases
- human robot interaction
- data driven
- reinforcement learning
- human robot
- human centered
- gesture recognition
- humanoid robot
- manipulation tasks
- optical flow
- pointing gestures
- light intensity
- service robots
- model free
- function approximation
- markov decision processes
- multi modal
- state space
- temporal difference
- natural interaction
- robot programming
- machine learning
- multi agent
- learning algorithm
- action selection
- reinforcement learning algorithms
- dynamic programming
- image sequences
- artificial intelligence
- qualitative spatial