Reconfigurable Control Allocation Technology Using Weighted Least Squares for Nonlinear System in Unmanned Aerial Vehicle

Abstract

Reconfigurable control allocation research is important to multidisciplinary science and engineering applications. In particular, the proposed research plays a significant role in enhancing the safety, reliability and fault tolerance capability of Unmanned Aerial Vehicle (UAV), which is one of the most active research and development areas. The main objective of this paper is to introduce and evaluate UAV reconfigurable control system design against control surfaces faults without modifying the baseline controller. The faults introduced are in the form of partial loss and stuck at unknown position on the UAV control surfaces. Weighted Least Squares (WLS) control reallocation algorithm with application to UAV was investigated. The paper is undertaken in a nonlinear UAV model ALTAV (Almost-Light-Than-Air-Vehicles), developed by Quanser incorporation. Different faults have been introduced in control surfaces with different trajectory commanded inputs. Gaussian noise was introduced in the model. Comparisons were made under normal situation, the case without control reallocation, and the case with control reallocation method. Simulation results show the satisfactory reconfigurable flight control system performance using the WLS control reallocation method for ALTAV nonlinear UAV benchmark.