Human Supervision of Multiple Autonomous Vehicles

摘要

To support the vision of a system that enables a single operator to control multiple next-generation unmanned air vehicles from a single workstation, a new interface paradigm is required. The interface must consider different types and levels of human-automation interactions. For instance, the interaction will move from one of human-directed mode selection and activation to more dynamic, bilateral and mixed-initiative human-automation decision- making, collaboration, negotiation, even conflict resolution across a range of situations. This project involved several threads of research to start examining human-automation interaction issues. One research thread focused on determining the appropriate level of automation for tasks envisioned for supervisory control of multiple UASs. Another thread involved the design and evaluation of a temporal interface for support of multi-UAS control and interaction with automation. Finally, a third research thread involved support for other projects that informed future operator-automation interfaces. This report provides a brief review of these research threads.