A real‐time field experiment on search and rescue operations assisted by unmanned aerial vehicles

Abstract

This paper reports on the performance of a novel system for supporting search and rescue activities, known as SARUAV (search and rescue unmanned aerial vehicle), in a field experiment during which a real-world search scenario was simulated. The experiment took place on March 2-3, 2017, at two sites located in southwestern Poland. Three groups acted in the experiment: (1) SARUAV and unmanned aerial vehicle (UAV) operators, (2) ground searchers, and (3) participants who simulated being lost. In the uncomplicated topography without snow cover, the system identified the lost persons, and ground searchers found them 31 min after the SARUAV report had been disseminated. In the mountainous area covered with snow, one person was found within 9 min after searchers received the SARUAV report; however, the other two persons were not identified by SARUAV. The field experiment served as a proof of concept of the SARUAV system, confirmed its potential in person identification studies, and helped to identify numerous scientific and technical problems that need to be solved to develop a mature version of the system. K E Y W O R D Slost person, nested k-means, ring model, target detection, unmanned aerial vehicle

Can Clouds Improve the Performance of Automated Human Detection in Aerial Images?

“…• A flight plan over a test area should be prepared so that a UAV operates at altitudes which are suitable for person detection in aerial images (Goodrich et al 2008;Niedzielski et al 2018) and flies above persons who pretend being lost in the wilderness. • A specific day should be chosen so that changes in vegetation do not occur, but the amount of cloud cover varies over this day-ideally with and without cloud impact on visibility of terrain from the UAV.…”

“…Rudol and Doherty 2008;Flynn and Cameron 2013) are carried out. In the current state of the UAV-based SAR technologies, there is a general consensus that the automated procedures may support human-assisted searches, but cannot fully substitute an expert who operates a sensor (Molina et al 2012;Niedzielski et al 2018).…”

“…Clouds, fog or smoke may impact performance of different cameras, and therefore the presence of such barriers is treated as a disadvantage in remote sensing (Oakley and Satherley 1998;Woodell et al 2015;Gultepe et al 2009;Djuricic and Jutzi 2013). In the context of UAV-assisted SAR work, bad weather not only influences image clarity but also limits the effectiveness of search through constraining flights and other field activities (Sumimoto et al 2000;Niedzielski et al 2018). However, the negative impact of deteriorated image clarity on the usability of such imagery can be reduced by data processing techniques (Steinvall et al 1999;Woodell et al 2015;Yuan et al 2010).…”

Unmanned aerial system–assisted wilderness search and rescue mission

“…Real-time onboard video processing may help to identify the most impacted areas and to assess whether any individuals need help. 5,7 The UAs can also assist the rescue teams to reach a particular location if standard routes are blocked.…”

Detection of bodies in maritime rescue operations using unmanned aerial vehicles with multispectral cameras

“…More specifically, UAVs and multispectral cameras are usually used for search and rescue tasks of lying bodies outdoors (Rudol & Doherty, ) where detection and localization are performed to train rescue routines in gravel roads, asphalt, and grass. Missing and lost people (standing and lying) can also be searched for by emergency services in natural environments such as uncovered terrain and forest (Niedzielski, Jurecka, Miziński et al, ; Niedzielski, Jurecka, Stec, Wieczorek, & Miziński, ), and in desert and chaparral (Coulter et al, ).…”