Diagnosis of Icing and Actuator Faults in UAVs Using LPV Unknown Input Observers

Abstract

This paper proposes a discrete-time linear parameter varying (LPV) unknown input observer (UIO) for the diagnosis of actuator faults and ice accretion in unmanned aerial vehicles (UAVs). The proposed approach, which is suited to an implementation on-board, exploits a complete 6-degrees of freedom (DOF) UAV model, which includes the coupled longitudinal/lateral dynamics and the impact of icing. The LPV formulation has the advantage of allowing the icing diagnosis scheme to be consistent with a wide range of operating conditions. The developed theory is supported by simulations illustrating the diagnosis of actuator faults and icing in a small UAV. The obtained results validate the effectiveness of the proposed approach. Keywords Unknown input observers (UIOs) • Linear parameter varying (LPV) systems • Icing diagnosis • Fault diagnosis • Unmanned aerial vehicles (UAVs) This work has been supported by the Research Council of Norway through the Centres of Excellence funding scheme (ref. 223254-AMOS). Damiano Rotondo is also supported by the ERCIM Alain Bensoussan Fellowship programme. This work has also been partially funded by the Spanish Government (MINECO) and FEDER through the project CICYT HARCRICS (ref. DPI2014-58104-R).

Reachability and stabilization of scheduled steady-states for LPV single-input systems

“…which is in turn equivalent to (24). Finally, if there exists a solution ρ(θ) to (20) which does not depend onθ, such solution makes v(θ) in (11) a RSP under the input u v given by (25).…”

“…For instance, among many other applications, one can list: aviation [29,36], robotic manipulators [15], automotive [12], motors [16], switched systems [4], mobile robots [27] and biomedicine [5]. In addition to controller design, several theoretical problems of interest have been solved within the LPV framework, such as filtering [3,21], fault detection [24,25,35] and fault tolerant control [26,22].…”

Actuator Faults Estimation for a Helicopter UAV in the Presence of Disturbances

“…UIO has been used to detect actuator saturation fault in the presence of disturbance in a helicopter with three degrees of freedom in (Lan et al 2017). Rotondo et al (2018) describe the discrete-time linear parameter varying UIO in UAVs. A sliding mode observer has been designed for detection, isolation and estimation of the actuator faults of a quadrotor with nonlinear Lipschitz model (Zheng et al 2017;Zhang et al 2013).…”

H−/L∞ fault detection observer for linear parameter‐varying systems with parametric uncertainty

“…As a result, LPV system has received much attention in fault diagnosis area due to its powerful capability of modeling nonlinear complex systems. () For example, a fault detection method based on sliding‐mode observer was proposed for uncertain LPV systems in the work of Chandra et al By using unknown input observer, Rotondo et al dealt with actuator fault diagnosis for UAVs. Rotondo et al proposed an interval unknown input observer to solve fault detection and isolation for uncertain LPV systems.…”