Loïc Lavigne

Nonlinear Observer-Based OFC Detection for A380 Aircraft

Abstract In this paper, we address the problem of Oscillatory failure Case (OFC) detection in A380 aircraft. The importance of OFC detection is well known in the literature because of natural mode excitation and the structural loads they can provoke. A novel nonlinear observer-based approach, which has two different gains is proposed in this paper to solve the problem. The main advantage of the proposed residual generator is that, by an adequate selection of observer gains, the residual generator has a kind selective (frequency dependent) attenuation property: disturbances and unmodelled dynamics effects on the residual are reduced while the effect of the fault is slightly amplified in a specific fault frequency range. Further, if the gains are chosen to be zero, the particular case of the technique implemented on A380 Airbus is obtained. A test based on real measurements of an A380 airplane is presented to show the efficiency of the proposed residual over the existing techniques.

Oscillatory Failure Case detection for new generation Airbus aircraft: A model-based challenge

Robust and early detection of Oscillatory Failure Case (OFC) in the Electrical Flight Control System (EFCS) of new generation aircraft (A/C) appears to be a challenging problem. OFC leads to strong interactions with loads and aero-elasticity and consequently must be detected in time. A robust analytical redundancy-based technique implemented in A380 Flight Control Computer (FCC) is used for detecting such unauthorized oscillatory events. The technique has been successfully validated and provides a complete OFC coverage without false alarms in the A380 EFCS. However, for upcoming and future generation A/C, it could be required to detect OFC with less important amplitude. To meet this requirement, it becomes necessary to get more sensitive fault indicating signals. It is shown that the model quality can be significantly improved by reliable estimating of some physical parameters. The fault indicating signals obtained with the proposed methodology are compared to those obtained from A380 FCC during flight tests. The results are quite encouraging and suggest that OFC with less important amplitude could be successfully detected by the new strategy.

Robust and Early Detection of Oscillatory Failure Case for New Generation Airbus Aircraft

This paper addresses the problem of Oscillatory Failure Case (OFC) detection in the Electrical Flight Control System (EFCS) of the Airbus airplanes. The work describes the status of on going research activity undertaken within a collaborative project between Bordeaux University (France) and Airbus. A hydraulic actuator model is presented, which is currently used as the basis for a robust analytical redundancy-based technique implemented in A380 Flight Control Computer (FCC) for detecting unauthorized oscillatory events. The technique has been successfully validated and provides a complete OFC coverage without false alarms in the A380 EFCS. In the context of aircraft overall optimization, for upcoming and future programs, a need has emerged recently to improve the detection performance in case of less important amplitude OFC that may occur in the servo control loops of actuators. To meet this requirement, it becomes necessary to model more efficiently the dynamical behavior of actuators to get more sensitive fault indicating signals. It is shown that the model quality can be significantly improved by reliable estimating of some physical parameters. The fault indicating signals are compared to those obtained from A380 computers during flight tests. The results are quite encouraging and suggest that OFC with less important amplitude could be successfully detected by the new strategy.

Flat Stellar Dynamic Filter for Gyroless Attitude Estimation

In this paper, we present an innovative attitude estimation filter based on differential flatness of the satellite dynamics with Rodriguez Modified Parameters for star tracker only Attitude and Orbit Control System (AOCS) architectures. The main characteristic of this estimator is to be able to maintain the satellite Absolute Pointing Error (APE) within safe limits, avoiding safe hold mode recovery in case of long duration loss of measurements (attitude control with no measurements for 1000s). This filter has been selected for in flight testing on PARASOL CNES microsatellite, and will be experimented by May 2013.

Flatness Approach to LFT Modelling

In this chapter, an LFT model representing the longitudinal dynamic of the gap between the open loop nominal HIRM+ model and the perturbed model on a trajectory is given. Five of the most relevant uncertainties on the pitch axis are considered. This model describes the gap along a trajectory passing through flight condition FC1, with an angle of attack equal to 6°. The proposed approach is based on a simplified longitudinal model used to determine a nominal trajectory and corresponding input. The specific outputs are angle of attack α and pitch angle θ. These outputs allow a parametrisation of state-space trajectory. First the LFT generation by flatness approach is described. Then an LFT model of the open loop HIRM+ model for the above flight condition is given. Thereafter an LFT model of the closed loop HIRM+RIDE is proposed to allow the identification of worst case stability margin.

Modeling of Disturbed Flat System for Robust Control Design

Abstract This paper aims at presenting a modelization of the uncertainty on the linear model obtained from the so-called linearizing feedback of dynamic flat systems. After such a linearizing feedback is computed to ensure good nominal tracking performance, a compact set of models can be computed by taking account of both state space disturbances and parametric errors. A robust linear feedback is then designed in order to guarantee a specified level of performance. This representation of the compact set of models makes use of the Linear Fractional Transformation. so that the global robustness of tracking performance is assessed by μanalysis.

A Genetic Fuzzy Logic Based Approach to Solving the Aircraft Conflict Resolution Problem

We use a genetic fuzzy logic approach for solving the aircraft conflict resolution problem. We consider a small uncertainty in the velocity and the maneuver parameters which causes each aircrafts position at any instant to be within a region of uncertainty represented by a convex hull. The objective is to find conflict-free trajectories for the aircraft that minimize the cost of maneuvers. This paper introduces our unique architecture that consists of a hidden layer of neurons and layer of Fuzzy Inference Systems (FISs). An artificial intelligence called EVE is used to train the system and once it is trained, its capability is evaluated on a set of test scenarios. We compare the cost and the computational time of our approach with that obtained by directly applying Genetic Algorithm (GA). The results show the effectiveness of our approach in finding quick near-optimal solutions.

Sequential Probability Ratio Test Using First Laplace Distribution for Oscillatory Fault Detection of an Hydraulic Actuator

Abstract This paper proposes a sequential test for fault detection in the case of Laplace distribution of residual. This approach is adapted from Walds sequential test and the paper establishes specific test thresholds for Laplace distributions. The suggested algorithm is applied to the detection of oscillatory failure cases on an Airbus A380 elevator actuator characterized by Laplacian distribution of residual. Moreover the proposed application shows that computing workload resulting from the proposed sequential test is similar to the classical Wald test.

LFT representation of a longitudinal perturbed aircraft model by flatness approach

In this paper, an linear fractional transformations (LFT) model representing, the dynamic of the gap between a nominal model and a perturbed model of a longitudinal aircraft fighter along a trajectory is given. The proposed approach is based on a simplified model used to determine a nominal trajectory and corresponding input. This simplified nonlinear model is a flat nonlinear system. As the measures on the so-called flat outputs, used in the linearizing feedback, are corrupted with noise and effects of unmodelled dynamics or parametric errors (model simplifications), the state space trajectory of the actual plant is not identical to the reference state space trajectory. Then the exact linearization is not achieved, but, as long as the tracking error remains small enough, the actual quasi-linearised plant can be modelled as a disturbed linear plant. The gap behaviour is modelled by a set of linear models described by a LFT model.