Further analysis of a multi-rotor drone for radar applications

Abstract

This study focuses on the analysis of an ARF-MikroKopter OktoXL drone manufactured by MikroKopter. This internship is part of the SIERA project, co-supervised by the Automatic Control Department of CentraleSupélec in cooperation with the SONDRA Laboratory. The goal of the SIERA project is to use the drone to obtain radar images of a certain ground. To do so, an accurate control of the attitude, and most importantly, speed, is needed. The drone study was initiated in 2014 with the CEI team focusing on a simplified model of the drone. In 2015, two CEI groups continued to work on the project, achieving the following points: · developing a complete model of the drone implemented via a simulator in MATLAB/Simulink; · developing some control laws tested on the proposed models; · developing a model for the air drag; · reviewing the source-code used by Mikrokopter in the ARF processor. The aim of this bachelor thesis is to provide a better understanding of the source-code that can help future CEI implement the different control laws. Prior to that, though, the model has to be validated. In this internship, some guidance and tools are proposed to work towards this validation. In a first instance, the hardware composing the Oktokopter is reviewed in Chapter III. Then the source-code provided on the SVN repository is further studied in Chapter IV, with a focus on the measurements and the FlightCtrl module. In Chapter V, the communication between the FlightCtrl and the BLCtrl is analyzed, which could help model validation and further understanding of the units and gains of the signals involved in the control process.. Additionally, appendix 5 reviews the tests performed by the CEI groups and during this internship to tune the air drag model developed by CEI 2.