City Research Online

Abstract

Deploying autonomous agents in safety critical domains, such as emergency evacuations, requires them to reason about and predict the often uncertain behaviour of humans. This paper presents Yanapay, a simulation toolkit designed to evaluate the decision-making strategies of autonomous Search and Rescue (SAR) agents operating in a transportation hub. Yanapay applies social psychology principles to simulate complex human behaviours, such as group dynamics and social identity. The toolkit enables researchers to design and benchmark agent architectures that reason about these social factors and predict human intent, for instance, which civilians are likely to aid in evacuation efforts. Yanapay automates the simulation workflow, allowing researchers to systematically evaluate and compare the performance of different strategies across diverse scenarios. A video demonstrating the tool is available at https://youtu.be/erLsxmFtHzs.

Publication Type:	Conference or Workshop Item (Paper)
Additional Information:	© 2026 International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). https://doi.org/10.65109/YQLN3900. This work is licensed under a Creative Commons Attribution International 4.0 License.
Publisher Keywords:	Agent-Based Simulation; Emergency Evacuation
Subjects:	B Philosophy. Psychology. Religion > BF Psychology H Social Sciences > H Social Sciences (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments:	School of Science & Technology
SWORD Depositor:	Symplectic Administrator

Export

Downloads