User profiling and behavioral adaptation for HRI: A survey

To effectively collaborate with people, robots are expected to detect and profile the users they are interacting with, but also to modify and adapt their behavior according to the learned models. The goal of this survey is to focus on the perspective of user profiling and behavioral adaptation. On the one hand, human-robot interaction requires a human-oriented perception to model and recognize the human actions and capabilities, the intentions and goals behind such actions, and the parameters characterizing the social interaction. On the other hand, the robot behavior should be adapted in its physical movement within the space, in the actions to be selected to achieve collaboration, and by modulating the parameters characterizing the interaction. In this direction, this survey of the current literature introduces a general classification scheme for both the profiling and the behavioral adaptation research topics in terms of physical, cognitive, and social interaction viewpoints.