Interaction acceptance modelling and estimation for a proactive engagement in the context of human-robot interactions

Abstract

Understanding human behavior in social environments provides valuable insights and information. When individuals require interaction with others, they rapidly assess the likelihood of engagement based on social signals and the displayed activity of the potential partner of interaction. We refer to this cognitive process as the Interaction Acceptance Belief (IAB). The concept of IAB finds application in various social robotic scenarios, including service tasks, proactive approaches, and reactive methods. In this paper, we present a comprehensive definition of Interaction Acceptance Belief and propose a methodology for its realistic modeling within real-world scenarios. Our approach aims to enhance the capabilities of social robots to effectively infer and adapt to human preferences, leading to efficient human-robot interactions. By conducting experimental evaluations, we establish the feasibility of developing a model that captures and represents the Interaction Acceptance Belief within a specific social context.