Presentation
Shared Mental Models for Unforeseen Events in Human-Robot Teams
DescriptionAs robots are increasingly integrated into unstructured environments, there is growing interest in improving human-robot team (HRT) resilience to unforeseen events. Traditional methods like teleoperation and contingency planning often struggle with cognitive overload and a reliance on prior knowledge, which may not always be available. This study investigates how different types of Shared Mental Models (SMMs), General (GSMM) versus Specific (SSMM), influence HRT performance during unforeseen events. A user study simulated a search and reconnaissance mission where teams faced an unforeseen obstacle. Results showed that teams with a GSMM, emphasizing broad principles rather than specific use cases, performed significantly better when navigating unforeseen events compared to SSMM teams. Additionally, team adaptability independently enhanced performance and was important during unforeseen events. These findings suggest that fostering GSMMs in HRTs can better prepare teams for unstructured environments. The work highlights the importance of embedding adaptability and generalizable knowledge into SMMs to enhance HRTs operational effectiveness in unstructured environments.
Event Type
Lecture
TimeTuesday, October 14th5:10pm - 5:30pm CDT
LocationGrand B
Human AI Robot Teaming (AI)