Autonomous Capabilities for Small Unmanned Aerial Systems Conducting Radiological Response: Findings from a High‐fidelity Discovery Experiment. Issue 4 (27th February 2014)

• Record Type: Journal Article
• Authors: Duncan, Brittany A.; Murphy, Robin R.; Kendoul, Farid; Siegwart, Roland; Roberts, Jonathan
• Is Part Of: Journal of field robotics. Volume 31:Issue 4(2014:Jul./Aug.)
• Page Start: 522
• Page End: 536
• Publication Date: 2014-02-27
• Subjects: Robots, Industrial -- Periodicals; Automatic control -- Periodicals; 629.892
• DOI: 10.1002/rob.21503
• Languages: English
• ISSNs: 1556-4959

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>This article presents a preliminary work domain theory and identifies autonomous vehicle, navigational, and mission capabilities and challenges for small unmanned aerial systems (SUASs) responding to a radiological disaster. Radiological events are representative of applications that involve flying at low altitudes and close proximities to structures. To more formally understand the guidance and control demands, the environment in which the SUAS has to function, and the expected missions, tasks, and strategies to respond to an incident, a discovery experiment was performed in 2013. The experiment placed a radiological source emitting at 10 times background radiation in the simulated collapse of a multistory hospital. Two SUASs, an AirRobot 100B and a Leptron Avenger, were inserted with subject matter experts into the response, providing high operational fidelity. The SUASs were expected by the responders to fly at altitudes between 0.3 and 30 m, and hover at 1.5 m from urban structures. The proximity to a building introduced a decrease in GPS satellite coverage, challenging existing vehicle autonomy. Five new navigational capabilities were identified: <italic>scan</italic>, <italic>obstacle avoidance</italic>, <italic>contour following</italic>, <italic>environment‐aware return to home</italic>, and <italic>return to highest reading</italic>. Furthermore, the data‐to‐decision process could be improved with autonomous data digestion and visualization capabilities. This article is expected to contribute to a better understanding of autonomy in a SUAS, serve as a requirement document for advanced autonomy, and illustrate how discovery experimentation serves as a design tool for autonomous vehicles.</p> </abstract>