Flight Vibration Testing of the T-FLEX UAV using Online Modal Analysis

Abstract

Flight testing of the UAV demonstrator T-FLEX in the EU project FLIPASED has provided a platform to demonstrate the capabilities of an online modal identification system based on miniaturized hardware. The system uses data from Micro Electro Mechanical System MEMS sensors via a real time interface to the Flight Control Computer FCC. The signal processing, modal analysis and mode tracking using machine learning are performed using state of the art algorithms in Python and run on the Onboard Computer OBCII which is a Raspberry Pi 4. The data is encoded and transmitted via radio frequency RF telemetry to a ground station, where it is decoded and plotted in an interactive GUI. The natural frequencies and damping ratios can be visualized as a function of time, Mach number or altitude. This allows engineers on the ground to assess the flutter stability of the aircraft in real time during flight testing. A flight test campaign at the DLR airport Cochstedt served as the first deployment of the flutter monitoring system. The system was demonstrated to function robustly without bugs or system failures and was capable of running in real time. The modal parameters were identified with high certainty and could be accurately tracked throughout the flight envelope. The integration of the online modal analysis system with the onboard flight control system for active flutter control will be the next logical progression of the developed system.