Fast Hildreth-based Model Predictive Control of Roll Angle for a Fixed-Wing UAV

Victor Truong Thinh Lam, Abdul Sattar, Liuping Wang, Mircea Lazar

Research output: Contribution to conferencePaperAcademic

Abstract

In this paper we consider Model Predictive Control (MPC) design for roll angle control for a Fixed-Wing Unmanned Aerial Vehicle (UAV) with Multiple Segmented Control Surfaces. The challenge of roll angle control for a Fixed-Wing UAV consists of switching between inner and outer aileron-pairs with hard constraints due to safety, energy saving and switching actuators. The novelty consists of formulating a hybrid-control problem as a switched linear constrained MPC-QP problem and switched state observer design for Fixed-Wing UAV. A fast novel QP-solver based on the active-set QP-solver Hildreth is developed to meet the real-time implementation target, which is to stay below the sampling time of Ts = 10 ms. The designed MPC controllers are simulated using Matlab. Simulations and the CPU-time from the improved QP-solvers show MPC to be a very good choice for real-time roll angle control for Fixed-Wing UAVs.
Original languageEnglish
Number of pages8
Publication statusPublished - Jul 2020
Event21st IFAC World Congress 2020 - Berlin, Germany
Duration: 12 Jul 202017 Jul 2020
https://www.ifac2020.org/

Conference

Conference21st IFAC World Congress 2020
CountryGermany
CityBerlin
Period12/07/2017/07/20
Internet address

Cite this