Using a Mini-UAV to Support Wilderness Search and Rescue: Practices for Human-Robot Teaming

摘要

Wilderness Search and Rescue can benefit from aerial imagery of the search area. Mini Unmanned Aerial Vehicles can potentially provide such imagery, provided that the autonomy, search algorithms, and operator control unit are designed to support coordinated human-robot search teams. Using results from formal analyses of the WiSAR problem domain, we summarize and discuss information flow requirements for WiSAR with an eye toward the efficient use of mUAVs to support search. We then identify and discuss three different operational paradigms for performing field searches, and identify influences that affect which human-robot team paradigm is best. Since the likely location of a missing person is key in determining the best paradigm given the circumstances, we report on preliminary efforts to model the behavior of missing persons in a given situation. Throughout the paper, we use information obtained from subject matter experts from Utah County Search and Rescue, and report experiences and "lessons learned" from a series of trials using human-robot teams to perform mock searches.