Towards Design Guidelines for Natural-Language-Based Human-Multi-Robot Collaboration in Domestic Environments

As domestic robots become prevalent, households will involve multiple robots with different capabilities, requiring people to coordinate and collaborate with several robots daily. This demands understanding key human–multi-robot interaction that shapes human experiences in everyday home activities. We investigate key dimensions of interaction using a VR role-playing study where humans and robots coordinate primarily through natural language during collaborative domestic tasks. We combine qualitative analysis of interaction and interview data with quantitative measures of coordination behavior to examine how three key dimensions shape collaboration and human experience: task-coordination dominance, robot autonomy, and robot personality. Our results reveal how decision authority shifts across tasks, how autonomy preferences vary with task demands and trust, and how robot personality can enhance engagement yet risk distraction without adaptation. Based on these findings, we distill design implications for building domestic multi-robot systems that better balance control, efficiency, and human experience.