Reconfigurable flight control using neural generalized predictive control

Soloway, Don; Haley, P.

doi:10.2514/6.2000-5328

Cited by 8 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to investigate the dynamics of a damaged aircraft and the application of control algorithms, NASA developed the Generic Transport Model (GTM). The GTM is a physics based simulation of a twin engine transport class generic aircraft, which is capable of simulating the effects of various damage cases on the aircraft dynamics [8]. The GTM serves as a platform for the application and development of MRAC and other adaptive flight control architectures.…”

Section: Integrated Resilient Aircraft Control Projectmentioning

confidence: 99%

Application of a Six Degree of Freedom Adaptive Controller to a General Aviation Aircraft

Lemon

2011

AIAA Guidance, Navigation, and Control Conference

View full text Add to dashboard Cite

In an effort to increase general aviation safety, a simplified model reference adaptive control (MRAC), adapting to modeling error with only a bias neuron, is applied to a desktop simulation of the Hawker Beechcraft Corporation (HBC) CJ-144 fly-by-wire aircraft in six degrees of freedom. MRAC has been experimentally applied to military and commercial aircraft by the National Aeronautics and Space Administration and the Department of Defense. Previous research at Wichita State University has demonstrated promising results for MRAC's application to general aviation in three degrees of freedom, both in its ability to achieve desired handling qualities and adapt to unexpected changes in aircraft dynamics. MRAC is used to control the aircraft in the nominal case and adapts to maintain control with changes in flight condition or unexpected A matrix and B matrix failures. The HBC CJ-144 aircraft is configured for decoupled flight control where the pilot commands the airspeed, flight path angle, bank angle, and side force. The controller architecture consists of an inverse controller, single bias neuron adaptive element, model follower and linear

show abstract

Section: Integrated Resilient Aircraft Control Projectmentioning

confidence: 99%

Application of a Six Degree of Freedom Adaptive Controller to a General Aviation Aircraft

Lemon

2011

AIAA Guidance, Navigation, and Control Conference

View full text Add to dashboard Cite

show abstract

“…A relatively novel, but very common approach is to use neural networks for adaptive control. Here, a variety of different architectures has been developed [15,5,2,7,20,19,10]. They differ substantially in the method of control (e.g., direct, inverse [2], predictive [15,19]), the neural network architecture (e.g., feed-forward networks [15], dynamic cell structures (DCS) [9,20], or SigmaPi [10]), as well as implementation details.…”

Section: Introductionmentioning

confidence: 99%

Toward V&V of neural network based controllers

Schumann

Nelson

2002

Proceedings of the First Workshop on Self-Healing Systems

View full text Add to dashboard Cite

Online adaptation is a powerful means to handle unexpected slow or catastrophic changes of the system's behavior (e.g., a stuck or broken rudder of an aircraft). Therefore, adaptation is one way for realizing a self-healing system. Substantial research and development has been made to use neural networks (NN) for such tasks (e.g., integrated in various unmanned helicopters and test-flown on a modified F-15 aircraft). Despite the advantages of adaptive neural network based systems, the lack of methods to perform certification, verification, and validation (V&V) of such systems severely restricts their applicability.In this paper, we report on ongoing work to develop V&V techniques and processes for NN-based safety-critical control systems, in our case an aircraft flight control system. Although the project ultimately aims at V&V of online adaptive systems, this paper focuses on the first part of this project dealing with so-called pre-trained neural networks (PTNN). V&V techniques developed here are important pre-requisites for handling the online adaptive case. In particular, we describe highlights of a process guide which has been developed within this project and discuss important V&V issues which need to be addressed during certification.

show abstract

“…The Newton-Raphson method is used as the optimization algorithm, which is shown to have a good com putational efficiency. In [95], authors report the result on reconfigurable flight control using NGPC. In both papers, there is a lack of stability proof.…”

Section: Introductionmentioning

confidence: 99%

Control of nonlinear flexible space structures

Shi

View full text Add to dashboard Cite

CHAPTER 1. INTRODUCTION 1 1.1 Background 1 1.2 Thesis outline 10 CHAPTER 2. APPROXIMATE MATHEMATICAL MODELS OF FLEX IBLE SPACE STRUCTURES 11 2.1 Introduction 11 2.2 Linear model of flexible single body system 13 2.2.1 Linear rigid body motion equations 2.2.2 Elastic motion equations 2.3 Approximate nonlinear model of flexible single body system 2.3.1 Euler parameters 2.3.2 State equations for flexible single body systems 21 2.4 Approximate nonlinear model of flexible multibody system 2.4.1 Coordinate systems 2.4.2 Kinetic energy 2.4.3 Lagrange equation and a fundamental equation for a system of rigid bodies 2.4.4 State equations for flexible multibody systems 2.5 Remarks V CHAPTER 3. STABLE GENERALIZED PREDICTIVE CONTROL ... 27 3.1 Introduction 27 3.2 Input/output approach for Generalized Predictive Control (GPC) 28 3.3 State-space approach for Generalized Predictive Control (GPC) 29 3.3.1 Stable generalized predictive control law 29 3.3.2 GPC control law with end-point state weighting 3.3.3 Computation of desired state trajectories 35 3.4 Q-synthesis methods 37 3.4.1 Fake Algebraic Riccati Equations (FARE) based Q-synthesis 3.4.2 Constrained nonlinear optimization based Q-synthesis 3.5 Reconfigurable control 3.5.1 Definitions 3.5.2 Stability under reconfigurable control 3.6 Numerical example 3.7 Remarks CHAPTER 4. CONTROL DESIGN 4.1 Introduction 4.2 Modal analysis 4.3 Stable reconfigurable GPC design 4.4 Nonlinear control design for single body systems 4.4.1 Linear GPC design 63 4.4.2 Feedback linearization based GPC (FLGPC) design 65 4.4.3 Nonlinear dissipative control design 77 4.4.4 Nominal performance comparison 82 4.4.5 Robust performance comparison 82 4.5 Nonlinear dissipative control design for multibody systems 96 4.5.1 Dynamical model verification 98 4.5.2 Nonlinear dissipative control design 99 vi 4.6 Remarks CHAPTER 5. CONCLUSIONS 107 5.1 Main research contributions

show abstract

Reconfigurable flight control using neural generalized predictive control

Cited by 8 publications

References 0 publications

Application of a Six Degree of Freedom Adaptive Controller to a General Aviation Aircraft

Application of a Six Degree of Freedom Adaptive Controller to a General Aviation Aircraft

Toward V&V of neural network based controllers

Control of nonlinear flexible space structures

Contact Info

Product

Resources

About