HCC: human-centered interfaces for safe and effective supervision of autonomous drone swarms

The National Science Foundation Division of Computing and Communication Foundations awarded the University of Notre Dame DU Lac $799,964 on October 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop human-centered interfaces for safe and effective supervision of autonomous drone swarms.

The project will create and evaluate interface designs and testing methods that enable teams of people to supervise autonomous drone swarms across simulation environments, remote operations, and real-world field deployments. The research combines cognitive work analysis, user-centered design, behavioral measurements, and biometric sensing to study how operators maintain awareness, coordinate with teammates, develop appropriate reliance on automation, and intervene when autonomous systems encounter unexpected situations. The project will establish a rigorously instrumented experimental platform for direct comparison of human performance across increasing levels of operational realism and will adapt automated stress-testing techniques to systematically discover failures that may emerge during human supervision of autonomous systems. The research addresses applications in disaster response, wildfire monitoring, infrastructure damage assessment, and environmental tracking, and will provide students with hands-on experience in human-centered design, autonomous systems, and field robotics.

Work is performed at the University of Notre Dame, Notre Dame, Indiana. The period of performance runs from October 1, 2026, through September 30, 2029.

Abstract

Autonomous drone swarms, groups of aerial vehicles that operate cooperatively without human controllers, could transform how society responds to disasters, monitors wildfires, assesses infrastructure damage, and tracks rapidly changing environmental conditions. In these situations, teams of people may need to oversee many autonomous drones simultaneously while making rapid decisions under uncertain and stressful conditions. However, current technologies often make IT difficult for people to understand drone behaviors, coordinate effectively with one another, and know when to step in before problems escalate. This project will develop human-centered interfaces that enable teams to safely supervise autonomous drone swarms in time-critical situations, improve the safety and reliability of autonomous technologies, and strengthen the nation's preparedness for future emergencies. The project will also provide students with hands-on research experience in human-centered design, autonomous systems, and real-world field robotics. The project will create and evaluate new interface designs and testing methods that support teams of people supervising autonomous drone swarms in simulation environments, remote operations, and real-world field deployments. The research combines cognitive work analysis, user-centered design, behavioral measurements, and biometric sensing to study how people maintain awareness, coordinate with teammates, develop appropriate reliance on automation, and intervene when autonomous systems encounter unexpected situations. The project will also establish a rigorously instrumented experimental platform that enables direct comparison of human performance across increasing levels of operational realism. A novel aspect of the research adapts automated stress-testing techniques to systematically discover failures that may emerge when people supervise autonomous drone swarms. Computational models of human behavior and uncertainty-aware evaluation methods will identify interaction problems that traditional user studies may overlook. These vulnerabilities will then be validated through experiments involving human supervisors and mitigated through iterative interface redesign. The resulting datasets, evaluation protocols, and interface designs will establish a scientific foundation for creating safe, transparent, and reliable methods for supervising future autonomous multi-agent systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are not planned for this award.