Toward Safe and Efficient Human–Robot Interaction via Behavior-Driven Danger Signaling

This article introduces the notion of danger awareness in the context of human–robot interaction, which decodes whether a human is aware of the existence of the robot, and illuminates whether the human is willing to engage in ensuring safety. This article also quantifies the notion as a single binary variable, called danger awareness coefficient, and provides a game-theoretic interpretation for that. Employing an online Bayesian learning method to update the robot’s belief about the human’s danger awareness by observing their actions, it is shown how the robot can build a predictive human model to anticipate the human’s future actions. To enrich robot’s observations, and thus to improve safety and efficiency of the robot, the robot is equipped with a danger signaling system to generate awareness in the human. Finally, a planning scheme is proposed to provide an efficient and probabilistically safe plan for the robot. The effectiveness of the proposed scheme is demonstrated through simulation studies on an interaction between a self-driven car and a pedestrian.