Loading...
Please wait, while we are loading the content...
Similar Documents
An on-line equivalent system identification scheme for adaptive control. ph.d. thesis - stanford univ.
| Content Provider | NASA Technical Reports Server (NTRS) |
|---|---|
| Author | Sliwa, S. M. |
| Copyright Year | 1984 |
| Description | A prime obstacle to the widespread use of adaptive control is the degradation of performance and possible instability resulting from the presence of unmodeled dynamics. The approach taken is to explicitly include the unstructured model uncertainty in the output error identification algorithm. The order of the compensator is successively increased by including identified modes. During this model building stage, heuristic rules are used to test for convergence prior to designing compensators. Additionally, the recursive identification algorithm as extended to multi-input, multi-output systems. Enhancements were also made to reduce the computational burden of an algorithm for obtaining minimal state space realizations from the inexact, multivariate transfer functions which result from the identification process. A number of potential adaptive control applications for this approach are illustrated using computer simulations. Results indicated that when speed of adaptation and plant stability are not critical, the proposed schemes converge to enhance system performance. |
| File Size | 4826241 |
| Page Count | 182 |
| File Format | |
| Alternate Webpage(s) | http://archive.org/details/NASA_NTRS_Archive_19840009117 |
| Archival Resource Key | ark:/13960/t4fn61081 |
| Language | English |
| Publisher Date | 1984-01-01 |
| Access Restriction | Open |
| Subject Keyword | Aircraft Stability And Control Algorithms Transfer Functions On-line Systems Computerized Simulation System Identification Adaptive Control Compensators Ntrs Nasa Technical Reports Server (ntrs) Nasa Technical Reports Server Aerodynamics Aircraft Aerospace Engineering Aerospace Aeronautic Space Science |
| Content Type | Text |
| Resource Type | Thesis |
