Integrated Physics-Data Based LPV Attitude Control of Quadrotor UAV System

This article proposes an integrated physics-data based attitude control method to achieve stable flight of a quadrotor unmanned aerial vehicle (UAV) system. First, a quasi-linear parameter varying (LPV) model with unknown parameter matrices is established to characterize quadrotor UAV nonlinear dynamics. Second, an integrated control approach fusing physical mechanisms and data information is developed. It derives an auxiliary physics-informed data-driven system representation by embedding a limited set of measured outputs and control inputs into the quasi-LPV formula with unknown parameter matrices. Then, a gain-scheduled feedback control law is designed based on this representation. The controller gains as well as control inputs are obtained directly using informative data that satisfy persistently exciting (PE) conditions, instead of through explicit model identification. Furthermore, the closed-loop system stability and tracking error convergence are mathematically proved based on Lyapunov theory, and the control gains are calculated through solving a semi-definite programming (SDP) problem with a set of linear matrix inequality (LMI) conditions. Finally, based on a quadrotor UAV platform, both simulations and experiments are performed to verify the effectiveness and performance of the proposed method.