Optimal tuning for robust control of a small fixed-wing UAV

An hybrid methodology, that combines the use of a robust servomechanism linear quadratic regulator (RSLQR) and a genetic algorithm for the design of the flight control system (FCS) of a lightweight unmanned aerial vehicle (UAV), is the subject of this paper. The goal is to devise a systematic design approach so as to reduce the burden of trial and error procedures while providing improved flight performance with respect to a classic FCS featuring PID inner loops on attitude angles for stability and control augmentation. FCS design is formulated as an optimization problem aimed at maximizing a weighted sum of suitably defined performance metrics evaluated through a number of nonlinear simulations, so as to stress the control system on specific maneuvers, like combined altitude and course variations at different flight speeds. Performance of the proposed control design methodology is assessed by using analytical tools for linear systems and software-in-the-loop (SIL) simulations of the UAV model.