Camera-Equipped Mini UAVs for Wilderness Search Support

Wilderness Search and Rescue (WiSAR) entails searching over very large regions in often rugged remote areas. Because of the large regions and the potential danger to ground searchers, WiSAR is an ideal application for using small (human-packable) UAVs to provide aerial imagery of the search region. In this paper, we use goal-directed task analysis to identify a set of requirements and to design algorithms for autonomous search patterns and operator interfaces for effective search. We then present results from flight tests, experiments with human participants, and search trials that employ these algorithms. These results provide evidence that effective search can be systematically performed by algorithmically placing waypoints, that detection rates increase when image mosaics are used, and that small UAVs can potentially be beneficial to real searches.