Quantifying Loss-Of-Control of Quadrotors

Abstract

With most of the current research in quadrotor Loss-Of-Control (LOC) being focussed on specific failure cases e.g. sensor faults and Single-Rotor Failure (SRF), the growth that is expected in the drone industry will not be able to be sustained, in regards to the safety of individuals in urban areas. Without an assurance of reliability regarding the safety of drones this is just not feasible. With the National Aeronautics and Space Administration (NASA) outlining flight traffic rules for drones it seems to be just a matter of time until it will be normal to see such vehicles flying around. Therefore it is of the utmost importance to improve the overall resilience of quadrotors. This work seeks to show the importance of modelling hazards such as the Vortex Ring State (VRS) and blade flapping to broaden the approach on solving LOC of quadrotors. Through the adaptation of the definition of LOC of aircraft to quadrotors and a comparative analysis of quadrotor flights, of both the nominal and SRF configuration, a Quantitative LOC Definition (QLD) for quadrotors is created. This definition is then validated through the analysis of thrust stand measurements and quadrotor flights. Resulting in a measure for the identification of LOC events.