Real time degradation identification of UAV using machine learning techniques

Abstract : The usages and functionalities of Unmanned Aerial Vehicles (UAV) have grown rapidly during the last years. They are being engaged in many types of missions, ranging from military to agriculture passing by entertainment and rescue or even delivery. Nonetheless, for being able to perform such tasks, UAVs have to navigate safely in an often dynamic and partly unknown environment. This brings many challenges to overcome, some of which can lead to damages or degradations of different body parts. Thus, new tools and methods are required to allow the successful analysis and identification of the different threats that UAVs have to manage during their missions or flights. Various approaches, addressing this domain, have been proposed. However, most of them typically identify the changes in the UAVs behavior rather than the issue. This work presents an approach, which focuses not only on identifying degradations of UAVs during flights, but estimate the source of the failure as well.
Document type :
Conference papers
Complete list of metadatas

Cited literature [23 references]  Display  Hide  Download
Contributor : Matthieu Geist <>
Submitted on : Monday, November 20, 2017 - 3:47:18 PM
Last modification on : Monday, April 1, 2019 - 11:04:09 AM
Long-term archiving on : Wednesday, February 21, 2018 - 2:00:47 PM


Files produced by the author(s)


  • HAL Id : hal-01629680, version 1


Anush Manukyan, Olivares-Mendez Miguel, Holger Voos, Matthieu Geist. Real time degradation identification of UAV using machine learning techniques. International Conference on Unmanned Aircraft Systems (ICUAS), 2017, Miami, United States. ⟨hal-01629680⟩



Record views


Files downloads