IM HERE – Interaction Model for Human Effort based Robot Engagement
We present our novel engagement modeling framework IMHERE, designed to improve human-robot interaction by capturing relational dynamics using a formal effort-based approach. The model applies to human-human, human-robot, and robot-robot interaction, and offers a modular implementation supporting real-world robotics integration.
IM HERE – Interaction Model for Human Effort based Robot Engagement
Engagement is a central requirement for natural and adaptive human-robot interaction (HRI), yet it remains difficult to generalize across settings. Our recent publication introduces \IMHERE, a formal and generalizable interaction model built on the idea that engagement can be modeled through observable and interpretable effort between entities.
The model captures bilateral relationships between any two entities—whether human, robot, or object—by tracking their focus, the allocation of cognitive or perceptual resources, and interpreting their emitted effort. Each interaction is reduced to one of four fundamental states: Passive, Requested, Buildup, and Engaged.
Unlike traditional models, \IMHERE bridges subjective and objective perspectives, includes mechanisms to identify miscommunication, and introduces principles for social norm-compliant interaction. The model applies equally to individual and group interactions, integrating concepts like F-formations and enabling social robots to reason about their role in dynamic, multi-party environments.
To support adoption, we provide an open-source simulator that enables testing and visualizing engagement in a variety of HRI scenarios.
Fulltext Access
The Paper is Accepted and will be published soon.
Citing
@inproceedings{strazdas2025imhere,
title={IM HERE: Interaction Model for Human Effort based Robot Engagement},
author={Strazdas, Dominykas and Jung, Magnus and Marquenie, Jan and Siegert, Ingo and Al-Hamadi, Ayoub},
year={2025}
}