Fault Adaptive Control Technology

   

Motivation and Goals

Current technology supporting integrated fault diagnostics and control in large-scale, heterogeneous systems is extremely limited. The objective of this research is to develop new techniques that will be able to:

  1. detect discrepancies between expected and observed behavior
  2. perform mode identification
  3. generate and verify fault hypothesis
  4. analyze the expected consequences of controller actions on system behavior
  5. derive system configuration and control law alternatives that maintain critical functionalities
  6. select new control law among the alternatives based on predicted transient behavior and performance
  7. manage reconfiguration  and take control actions to confine the fault and restore system operation with the resources available 

Project Scope

The project focuses on developing tools for the construction of dynamically reconfigurable control systems, which will adaptively reconfigure themselves based on diagnosis of faults. The tools will implement a domain-specific environment that is customized to the application arena.

The project includes 

  • A hybrid compositional modeling technology integrated with a model-integrated computing environment.
  • Monitoring, mode identification, diagnosis, and recovery control methods.
  • On-line controller synthesis and hardware reconfiguration
  • Transient management for construction of structurally adaptive controllers

The developed algorithms and tools will be demonstrated and evaluated using selected problems obtained from aerospace. 

Publications

Quadchart

Internal

Presentations:

This project is a DARPA funded effort.

This project is conducted in cooperation with the Department of Measurement and Information Systems, Technical University of Budapest, Hungary.

Software:

(Latest Release - 04/14/03. Previous release available)

(Latest Release - 01/06/03. Previous release available)

 

Contact:

For additional information, please contact Gabor Karsai at gabor@vuse.vanderbilt.edu or at (615) 343-7471.