Aerospace Control and Guidance Systems Committee

Announcements


You must first log in to access prior meeting presentations, register for a meeting, or nominate some for the Ward Award.


If you do not have a login account, or cannot remember the email address associated with your account, please click on the Application Form link below.

 
 

Login

 

E-mail: 

 

Password: 


Forgot your password?

Application Form


 

Site Search

Search our site:
 
 

Upcoming Events


Register for Meeting 132
(please log in first)

 
 

Photos


Meeting Highlights New!

Subcommittee S

 
 

Prior Meetings

Abstracts may be viewed by anyone. Presentations are only available to active members who have logged in.

Meeting 132
(coming soon)

Meeting 131

Meeting 130

Meeting 129

Meeting 128

Meeting 127

Meeting 126

Meeting 125

Meeting 124

Meeting 123

Meeting 122

Meeting 121

Meeting 120

Meeting 119

Meeting 118

Meeting 117

Meeting 116

Meeting 115

Meeting 114

Meeting 113

Meeting 112

Meeting 111

Meeting 110

Meeting 109

Meeting 108

Meeting 107

Meeting 106

Meeting 105

Meeting 104

Meeting 103

Meeting 102

Meeting 101

Meeting 100

Meeting 99

Meeting 98

Meeting 97

Meeting 96

Meeting 95

Meeting 94

Meeting 93

Meeting 92

 
HomeWard Memorial AwardPlanning Advisory BoardDownloadsConstitution and By-LawsAboutHistoryContact Us

  ← Return to agenda

MeetingACGS Committee Meeting 100 - Cocoa Beach - October 2007
Agenda Location5 SUBCOMMITTEE D--DYNAMICS, COMPUTATIONS AND ANALYSIS
5.2 Real-Time Stability Margins in Flight Test
TitleReal-Time Stability Margins in Flight Test
PresenterDave Ward
AffiliationBarron Associates
Available Downloads*presentation
*Downloads are available to members who are logged in and either Active or attended this meeting.
AbstractAs the complexity of flight controllers grows so does the cost associated with verification and validation (V&V). Current-generation controllers are reaching levels of complexity that push the envelopes of existing V&V approaches, and further increases in controller complexity are required to provide the operational capabilities desired for next generation systems. Without improved approaches, there is little hope for affordable V&V of next-generation intelligent systems, and thus little hope that they can be safely fielded. Barron Associates believes that successful V&V of advanced control approaches will require a combination of a variety of tools to address all aspects of the control-law design and testing process, as well as methods for monitoring the behavior of high-risk components in operational settings. The current work addresses the safety of the flight test program for advanced control laws. Barron Associates, under a NASA Langley SBIR contract, is developing a software tool that estimates stability margins in real-time during a flight test, and compares these to a priori predictions to validate the design model. The tool uses worst-case analysis techniques to develop off-line estimates of worst-case stability margins at various flight conditions based on a prior uncertain linear models corresponding to those flight conditions. On-line, “broken-loop” transfer functions are estimated using recursive, non-parametric system identification techniques, and margin estimates are derived from these transfer function estimates. In addition to providing efficient on-line computations, the selected estimation approach provides a quantification of the uncertainty in the estimate. This uncertainty in the transfer function estimate provides an indication of when sufficient data has been collected to draw useful conclusions. The team is also developing approaches for online fitting of simplified uncertainty models to observed data, and on-line worst-case margins can be computed from these models. The user is alerted to potential safety problems when on-line margins degrade below a minimum acceptable bound, and when margins degrade beyond off-line worst-case estimates, which indicates an invalid design model. Ultimately, the tool will provide safety estimates for the current test condition as well as planned future test points. The software tool is implemented using Matlab and Java, with Matlab used for backend computations and Java used to interface to a real-time data stream and to provide a graphical user interface.



Copyright © 2024 | Question? webmaster@acgsc.org