SensHRPS: Sensing Comfortable Human-Robot Proxemics and Personal Space With Eye-Tracking

2025-12-09

AI Summary

This study investigates comfortable interpersonal distance perception in humanoid robot interaction, grounded in peripersonal space norms. We recorded gaze behavior—including dynamic pupillary responses—using a mobile eye tracker from 19 participants interacting with a humanoid robot at varying distances, and concurrently collected subjective comfort ratings. A physiologically informed, interpretable comfort prediction model was developed. Results reveal that the physiological comfort threshold in human–robot interaction significantly differs from that in human–human interaction; minimum pupil diameter emerges as the most discriminative single physiological indicator. A decision tree model achieves superior performance (F1 = 0.73) over deep learning alternatives and provides transparent, rule-based interpretability. This work establishes the first modeling paradigm integrating oculomotor physiological signals with interpretable machine learning for comfort assessment. It offers both theoretical foundations and practical tools to enable adaptive distance regulation in humanoid robots.

Technology Category

• Intelligent Robots: Human-Robot Interaction
• Humans and AI: Interaction Techniques and Devices
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Social robots must adjust to human proxemic norms to ensure user comfort and engagement. While prior research demonstrates that eye-tracking features reliably estimate comfort in human-human interactions, their applicability to interactions with humanoid robots remains unexplored. In this study, we investigate user comfort with the robot "Ameca" across four experimentally controlled distances (0.5 m to 2.0 m) using mobile eye-tracking and subjective reporting (N=19). We evaluate multiple machine learning and deep learning models to estimate comfort based on gaze features. Contrary to previous human-human studies where Transformer models excelled, a Decision Tree classifier achieved the highest performance (F1-score = 0.73), with minimum pupil diameter identified as the most critical predictor. These findings suggest that physiological comfort thresholds in human-robot interaction differ from human-human dynamics and can be effectively modeled using interpretable logic.

Problem

• Investigates user comfort with humanoid robots using eye-tracking
• Evaluates machine learning models to estimate comfort from gaze features
• Identifies differences in comfort dynamics between human-robot and human-human interactions

Innovation

• Eye-tracking estimates human-robot comfort distances
• Decision Tree model outperforms deep learning for comfort prediction
• Minimum pupil diameter is key predictor for robot proxemics