Deep reinforcement learning-based safe interaction for industrial human-robot collaboration using intrinsic reward function.
Quan LiuZhihao LiuBo XiongWenjun XuYang LiuPublished in: Adv. Eng. Informatics (2021)
Keyphrases
- human robot
- reward function
- reinforcement learning
- action selection
- reinforcement learning algorithms
- human robot interaction
- markov decision processes
- state space
- dialogue system
- inverse reinforcement learning
- partially observable
- optimal policy
- hierarchical reinforcement learning
- humanoid robot
- transition model
- function approximation
- markov decision process
- temporal difference
- multiple agents
- model free
- learning algorithm
- gesture recognition
- initially unknown
- transition probabilities
- robotic systems
- dynamic programming
- human users
- learning agent
- data mining
- machine learning
- artificial life
- search space
- mobile robot
- computational intelligence
- evolutionary computation
- maximum entropy