Application of Robust Control and Gain Scheduling to Missile Autopilot Design.
Abstract
The problem of gain scheduling LPV systems is studied and applied to missile autopilot design. Necessary and sufficient conditions for determining the Quadratic stabilizing of LPV systems are given. A stabilizing LPV controller is found if one exists. The necessary and sufficient conditions are realistically computable with reduced dimension and with only the elements of the Lyapunov matrix P as the unknown variables. The quadratic Stability results are then extended to the more general case of General Lyapunov stability. The derivation is the same as for Quadratic stability, except there is an additional term that is a function of the derivative of the Lyapunov function. An algorithm is presented that will solve these new conditions. Finally two methods for finding minimal/near minimal realizations for 2-D systems is presented. The first method requires extracting any greatest common divisors out of a nonminimal realization and then returning the realization back to state-space form. This method does not extend to N-D systems. However the second method, the System Equivalent Based method, can be extended to N-D systems. This method begins with the LPV system description being converted to Rosenbreock's system matrix form. Then the system matrix form is systematically manipulated until it is in state-space form.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1998
- Accession Number
- ADA357847
Entities
People
- S. L. Fields
- T. E. Bullock
Organizations
- University of Florida