A Framework and Algorithm for Human-Robot Collaboration Based on Multimodal Reinforcement Learning

Table 3

The questionnaire.

Intentions	Body gestures that can express the intention (if none, fill in 0)	Hand gestures that can express that intention (if none, fill in 0)	Keywords that can express the intention (if none, fill in 0)
The robot stops immediately
The robot takes a block and gives it to the user
The robot actively picks up a block
The robot takes the block from the user
Put aside the block that the robot clips up
Put the blocks that the robot clips up onto the Jenga tower