| Real-world social environments, such as future urban spaces populated by delivery, service, and humanoid robots, will inherently involve complex multi-agent social interactions. However, modeling and designing such systems introduce substantial challenges. Technologically, robots must perform scalable perception, track and reason about multiple humans, coordinate with other robots, and make real-time decisions in dynamic settings. From a modeling perspective, social multi-human, multi-robot interactions involve interdependent behaviors, shared attention, role allocation, and emergent group dynamics that are difficult to formalize within existing frameworks. In this position paper, we address this gap by proposing a novel formulation of multi-human, multi-robot interaction in social contexts, allowing us to shed light on currently open challenges. Our theoretical framework captures the coupled dynamics among agents and provides a principled foundation for analysis, coordination, and evaluation. By formalizing this problem, we aim to advance HRI toward scalable, socially aware robotic ecosystems that better reflect the complexity of real-world environments. |
*** Title, author list and abstract as submitted during Camera-Ready version delivery. Small changes that may have occurred during processing by Springer may not appear in this window.