Responsibility and Engagement - Evaluating Interactions in Social Robot Navigation

2025-09-16

AI Summary

Existing conflict assessment methods in Social Robot Navigation (SRN) lack dynamic modeling and interpretability for human–robot interaction conflicts. Method: This paper proposes a time-normalized conflict evolution modeling framework, introducing the *Engagement* metric and extending the *Responsibility* framework to dynamically quantify each agent’s behavioral contribution during conflict resolution or escalation. Contribution/Results: The resulting Responsibility–Engagement joint evaluation system enables objective, interpretable assessment of robot behavior quality and proactiveness. Validated across binary interactions, group collaboration, and dense crowd simulation scenarios, the method effectively captures key temporal characteristics of cooperative conflict resolution. It demonstrates strong generalizability and interpretability, establishing the first systematic, scalable, and quantifiable tool for SRN behavioral evaluation.

Technology Category

• Intelligent Robots: Human-Robot Interaction
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

In Social Robot Navigation (SRN), the availability of meaningful metrics is crucial for evaluating trajectories from human-robot interactions. In the SRN context, such interactions often relate to resolving conflicts between two or more agents. Correspondingly, the shares to which agents contribute to the resolution of such conflicts are important. This paper builds on recent work, which proposed a Responsibility metric capturing such shares. We extend this framework in two directions: First, we model the conflict buildup phase by introducing a time normalization. Second, we propose the related Engagement metric, which captures how the agents' actions intensify a conflict. In a comprehensive series of simulated scenarios with dyadic, group and crowd interactions, we show that the metrics carry meaningful information about the cooperative resolution of conflicts in interactions. They can be used to assess behavior quality and foresightedness. We extensively discuss applicability, design choices and limitations of the proposed metrics.

Problem

• Evaluating human-robot interaction quality in social navigation
• Measuring responsibility shares in conflict resolution between agents
• Quantifying how agent actions intensify or resolve conflicts

Innovation

• Time normalization for conflict buildup modeling
• Engagement metric for conflict intensity assessment
• Responsibility metric extension for cooperative resolution