Optimal H#x221E; controller on the stability of MAVs in a novel Software-in-the-Loop control platform

摘要

This paper presents a novel Software-in-the-Loop (SiL) evaluation of an optimal H∞ robust controller on the stability problem of MAVs (Micro Aerial Vehicles) in the quadrotor configuration, whose originality is to employ ©Ascending Technologies Pelican MAV. The synthesis of the robust controller is grounded by the γ-iteration algorithm, which results in a MIMO optimal controller bounded by an attenuation level. The core of SiL platform, a customized C++/C# software named FVMS (Flight Variables Management System), developed by ART (Aerial Robots Team) at USP/EESC, is able to deal with full duplex communication to ©Microsoft Flight Simulator (MSFS). In turn, MSFS acts as the virtual environment through which a complete dynamic and graphic model of the Pelican MAV can be configured and emulated. Since FVMS can fully reach every variable related to the simulated MAV, the optimal H∞ control algorithm can be implemented for evaluation in SiL simulation. Hence, the stability of the Pelican MAV can be observed. Regarding MAVs control evaluation, SiL simulation potentially contributes to save battery time, to ease control synthesis and prototyping and to prevent accidents during tests with the real robot. Results of Pelican stabilization in SiL simulation in hovering mode are presented and discussed.