Modelling situation awareness relations in a multiagent system

Blom, Henk A. P.; Sharpanskykh, Alexei · 2015 · OpenAlex-citations

DOI: 10.1007/s10489-015-0651-4

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of modeling Situation Awareness (SA) in multiagent systems (MAS) representing complex sociotechnical systems. While SA is critical for agent-based modeling, existing literature offers conflicting definitions: SA as a process, a product, or a combination of both. Traditional agent-based models often treat agents as isolated information processors, adding social interactions as secondary features. The authors argue that social relations and SA dynamics must be central to the model. To resolve this, they develop a mathematical framework for Multiagent Situation Awareness (MA-SA) relations that accommodates all three theoretical views of SA. The methodology builds upon Endsley’s three-level SA model (perception, interpretation, projection) and Endsley and Jones’ extension for shared SA. The authors formalize a novel framework for a system of $N$ agents and non-agent entities. They define MA-SA relations ($S_{ik}$) as sets of pairs mapping state elements of one agent to the state elements of another, allowing for asymmetric awareness (e.g., Agent A knows Agent B’s state, but not vice versa). The framework distinguishes between an agent’s base state, self-awareness, and SA regarding other agents. It also formalizes SA update processes corresponding to Endsley’s levels: Level 1 involves observation or messaging functions; Level 2 involves interpretation functions that update SA based on observations/messages; and Level 3 involves projection functions that update the agent’s base state and self-awareness through prediction and planning. The model explicitly accounts for uncertainties and errors in these processes. The paper demonstrates the framework’s application through a retrospective analysis of the Überlingen mid-air collision. It illustrates how the model captures MA-SA differences and inconsistencies. Specifically, it distinguishes between known SA differences (discrepancies in shared requirements) and unknown SA differences (inconsistencies where an agent’s belief about a state differs from the true state beyond its confidence interval). The framework allows for the modeling of "Theory of Mind," where an agent maintains SA about another agent’s SA. The results show that the formalism can represent complex, asymmetric, and nested awareness relations, including cases where agents hold erroneous but shared beliefs. The significance of this work lies in providing a rigorous, unified mathematical foundation for modeling SA in MAS. By integrating social relations at the core of the agent model rather than as an add-on, the framework better captures the dynamics of sociotechnical systems. It supports the analysis of conflicts, shared understanding, and safety risks in domains like air traffic management. The approach enables systematic differentiation between self-awareness, awareness of others, and awareness of non-agent entities, offering a tool for designing and analyzing systems where human-agent and agent-agent interaction is critical.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• situation awareness theory
• situational awareness