Jong-Yeob Shin

Blending Methodology of Linear Parameter Varying Control Synthesis of F-16 Aircraft System

The design of a linear parameter varying (LPV) controller for the F-16 longitudinal axes over the entire e ight envelope, using a blending methodology that lets an LPV controller preserve regional optimal solutions over each parametersubsetand reducescomputationalcostsforsynthesizing anLPVcontroller,ispresented.Threeblending LPV controller synthesis methodologies are applied to control F-16 longitudinal axes. By the use of a function substitution method, a quasi-LPV model of the F-16 longitudinal axes is constructed from the nonlinear equations of motion over the entire e ight envelope, including nontrim regions, to facilitate synthesis of LPV controllers for the F-16 aircraft. The nonlinear simulations of theblended LPV controller show that the desired performance and robustness objectives are achieved across all altitude variations.

Performance analysis on fault tolerant control system

In a fault tolerant control (FTC) system, a parameter varying FTC law is reconfigured according to fault parameters estimated by fault detection and isolation (FDI) modules. FDI modules require some time to detect fault occurrences in aero-vehicle dynamics. In this brief, an FTC analysis framework is provided to calculate the upper bound of an induced-L2 norm of an FTC system in the presence of false identification and detection time delay. The upper bound is written as a function of a duration time interval and exponential decay rates and has been used to determine which FTC law produces less performance degradation (tracking error) due to false identification. The analysis framework is applied for an FTC system of a highly maneuverable aircraft technology (HiMAT) vehicle

Adaptive Linear Parameter Varying Control Synthesis for Actuator Failure

A robust linear parameter varying (LPV) control synthesis is carried out for a Highly Maneuverable Aircraft Technology (HiMAT) vehicle subject to loss of control effectiveness. The scheduling parameter is selected to be a function of the estimates of the control effectiveness factors. The estimates are provided online by a two-stage adaptive Kalman filter estimator. The inherent conservatism of the LPV design is reduced through the use of a scaling factor on the uncertainty block that represents the estimation errors of the effectiveness factors. Simulations of the controlled system with the online estimator show that a superior fault tolerance can be achieved.

Results on worst-case performance assessment

Rather than focus on computation of robust stability conditions for systems with real parametric uncertainty, we consider worst-case performance due to real parametric uncertainty. We focus on a constant matrix problem, and perform optimisation to find worst-case bounds.

Robustness Analysis of Integrated LPV-FDI Filters and LTI-FTC System for a Transport Aircraft

This paper proposes an analysis framework for robustness analysis of a nonlinear dynamics system that can be represented by a polynomial linear parameter varying (PLPV) system with constant bounded uncertainty. The proposed analysis framework contains three key tools: 1) a function substitution method which can convert a nonlinear system in polynomial form into a PLPV system, 2) a matrix-based linear fractional transformation (LFT) modeling approach, which can convert a PLPV system into an LFT system with the delta block that includes key uncertainty and scheduling parameters, 3) µ-analysis, which is a well known robust analysis tool for linear systems. The proposed analysis framework is applied to evaluating the performance of the LPV-fault detection and isolation (FDI) filters of the closedloop system of a transport aircraft in the presence of unmodeled actuator dynamics and sensor gain uncertainty. The robustness analysis results are compared with nonlinear time simulations. Nomenclature

Worst-Case Analysis of the X-38 Crew Return Vehicle Flight Control System

A linear fractional transformation (LFT) model of the linearized equation of the lateral-directional axes of the X-38 crew return vehicle is developed to facilitate the analysis of e ight control systems. The LFT model represents uncertainty in nine aerodynamic stability derivatives at a given e ight condition with frequency-domain performance specie cations. The X-38 LFT model combined with a controller at specie c e ight conditions is used to determine the aerodynamic coefe cients that result in the worst-case performance and gain/phase margin of the closed-loop system. The objective is to verify that a given controller remains stable and achieves desired performance objectives for all predee ned aerodynamic variations at select operating conditions along its e ight trajectory.

Uncertainty Modeling for Robustness Analysis of Control Upset Prevention and Recovery Systems

Formal robustness analysis of aircraft control upset prevention and recovery systems could play an important role in their validation and ultimate certification. Such systems (developed for failure detection, identification, and reconfiguration, as well as upset recovery) need to be evaluated over broad regions of the flight envelope and under extreme flight conditions, and should include various sources of uncertainty. However, formulation of linear fractional transformation (LFT) models for representing system uncertainty can be very difficult for complex parameter-dependent systems. This paper describes a preliminary LFT modeling software tool which uses a matrix-based computational approach that can be directly applied to parametric uncertainty problems involving multivariate matrix polynomial dependencies. Several examples are presented (including an F-16 at an extreme flight condition, a missile model, and a generic example with numerous crossproduct terms), and comparisons are given with other LFT modeling tools that are currently available. The LFT modeling method and preliminary software tool presented in this paper are shown to compare favorably with these methods.

Linear Parameter Varying Control Synthesis for Actuator Failure, Based on Estimated Parameter

The design of a linear parameter varying (LPV) controller for an aircraft at actuator failure cases is presented. The controller synthesis for actuator failure cases is formulated into linear matrix inequality (LMI) optimizations based on an estimated failure parameter with pre-defined estimation error bounds. The inherent conservatism of an LPV control synthesis methodology is reduced using a scaling factor on the uncertainty block which represents estimated parameter uncertainties. The fault parameter is estimated using the two-stage Kalman filter. The simulation results of the designed LPV controller for a HiMXT (Highly Maneuverable Aircraft Technology) vehicle with the on-line estimator show that the desired performance and robustness objectives are achieved for actuator failure cases.

Analysis of linear parameter varying system models based on reachable sets

Presents the analysis method of quasi-LPV models, comparing the ellipsoid set which contains the reachable set of a nonlinear system to determine which quasi-LPV model is less conservative to present the nonlinear dynamics. Quasi-LPV models are constructed from a nonlinear model using different methods, to facilitate synthesis of an LPV controller for the nonlinear system. The comparison results of the closed-loop system performance with the synthesized LPV controllers correspond to the analysis results of quasi-LPV models.

Optimal blending functions in linear parameter, varying control synthesis for F-16 aircraft

Presents a methodology of blending two linear parameter varying (LPV) controllers over the entire parameter spaces. Optimal blending matrix functions are calculated to preserve each performance level of the LPV controller synthesized over each parameter subspace. The design of an LPV controller for the F-16 longitudinal axes over the entire flight envelope is demonstrated using this blending approach. The nonlinear simulations of the blended LPV controller show that the desired performance and robustness objectives are achieved across all altitude variations.