System identification begins with measured aircraft motion and "inverts" the responses to rapidly extract a model that accurately reflects the measured aircraft motion. This is an advanced modeling tool ideal for engineering design and data analysis within the Software, Intelligent Systems and Modeling Program.
System identification is a procedure by which a mathematical description of vehicle or component dynamic behavior is extracted from test data. It can be thought of as an inverse of simulation. CIFER is a very advanced modeling tool that has been tested and used both in government and industry. Together with the skilled application of the Ames research staff, it can provide a vital service to any physics-based modeling effort conducted under the Human & Robotic Technology program. Whether it is the Crew Exploration Vehicle or aerial vehicles exploring Mars or other destinations, these Ames capabilities will be essential in analyzing and predicting vehicle behavior and preventing potential problems in a time-efficient and cost-effective manner.
The identification and analysis procedures described below have been implemented in a comprehensive set of user-oriented programs supported by extensive databasing and utilities. Such integrated software packages are critical in the application systems identification tools; CIFER® is the first such integrated package for the frequency response methodology.
Caption: Measuring aircraft performance.
Key features of the CIFER® approach are:
- Identification algorithms highly exercised and tuned based on many flight projects.
- Highly flexible and interactive definition of identification model structures.
- Fully automated weighting function selection based on frequency-response accuracy.
- Reliable parameter accuracy metrics.
- Integrated procedure for identification and model structure determination.
- Time-domain verification of models, including identification of offsets and biases.
Application modules within CIFER allow the:
- rapid identification of transfer-function models.
- spectral signal analysis.
- handling qualities and classical servoloop analysis.
- time and frequency-domain comparisons of identification versus simulation model predictions.
The CIFER® user interface was designed and implemented in such a way as to relieve the user of bookkeeping concerns. The program parameters comprising this case are presented to the user on a series of screens, through which the user steps, changing default and/or saved values as needed. When all the input choices are made for a particular program, the system submits a batch job to conduct the computations and generate any requested plot files. This allows rapid evaluation of alternative models and flight data.
Changes in the model definition are achieved by simply moving the cursor around on the user screens and changing the default or previously saved setups.
The user can update the database with the changes that are made. Utilities have been developed to allow quick inspection, searching, plotting, or tabulated output of the contents of the database. The extensive and integrated databasing in CIFER® is a key requirement for organizing and analyzing the large amounts of data that are generated for flight test identification projects.
CIFER® is a uniquely suited tool for comprehensive analysis of aircraft and component dynamics. The CIFER® system has been exercised on a wide range of rotary-wing and fixed-wing aircraft flight programs including: XV-15, Bell-214ST, BO-105, AH-64, UH-60, V-22, AV-8B Harrier, and OH-58D. Applications to analysis of simulations/simulators include the ASTOVL, LHX, and VMS.
Caption: A vital service to any physics-based modeling effort.
The CIFER® software operates on VMS and UNIX platforms. All graphics and computational libraries for CIFER® are self-contained and do not depend on access to proprietary packages. The final dependency is in the way time history data is made available to the software. Although the extensive parameter identification experience and intimate knowledge of the software is resident in the highly skilled Ames research staff, CIFER® can be licensed under the Federal Technology Transfer Act of 1986. A comprehensive 25-hour video of engineering lectures and computer demonstration on CIFER® has been developed to train new users.
The solutions to the most challenging problems in vehicle research start with model identification of a vehicle system. A researcher first needs to understand how a vehicle currently behaves in order to investigate problems, make improvements, and add supplemental systems successfully.
Researchers at Ames have jointly developed an integrated facility of skilled researchers and tools for system identification based on a comprehensive frequency-response approach that is uniquely suited to the difficult problems associated with time-domain and flight test data analysis. The foundation of the approach used by NASA researchers is the Comprehensive Identification from Frequency Responses (CIFER®) tool set. CIFER® performs the high-quality extraction of a complete multi-input/multi-output (MIMO) set of non-parametric input-to-output frequency responses. These responses fully characterize the coupled characteristics of the system without a-priori assumptions. Advanced Chirp-Z transform and composite optimal window techniques developed and exercised with over 10 years of flight project applications provide significant improvement in frequency-response quality relative to standard Fast Fourier Transforms (FFTs). Sophisticated nonlinear search algorithms are used to extract a state-space model that matches the complete input/output frequency-response data set.