EVALUATION TECHNIQUES FOR ADAPTIVE OPTICS CONTROL SYSTEMS

THIS PROPOSAL WILL REFINE RECENT STABILITY-ROBUSTNESS AND PERFORMANCE-ROBUSTNESS METHODS FROM MULTIVARIABLE FEEDBACK CONTROL THEORY AND USE THESE CONCEPTS TO DEVELOP PRACTICAL TESTING TECHNIQUES, AND EVENTUALLY TESTING HARDWARE, FOR EVALUATING PERFORMANCE PROPERTIES OF ADAPTIVE OPTICS SYSTEMS. THE KEY TO THE DEVELOPMENT OF THE TESTING METHODOLOGY AND HARDWARE IS THE EXPERIMENTAL CALCULATION OF THE SINGULAR VALUES VERSUS FREQUENCY OF RELEVANT (SENSITIVITY AND CLOSED-LOOP) TRANSFER FUNCTION MATRICES ASSOCIATED WITH THE ADAPTIVE OPTICS FEEDBACK CONTROL SYSTEM. AN AUTOMATED TESTING PROCEDURE INVOLVING AN INTEGRATED SET OF TEST HARDWARE AND SOFTWARE WILL BE DEVELOPED. THIS TEST HARDWARE WILL INJECT A VECTOR OF SINUSOIDS (AMPLITUDE AND PHASE) AT THE SYSTEM SENSORS FOR A RANGE OF FREQUENCIES. THE SOFTWARE WILL TRANSLATE THIS SINUSOIDAL INFORMATION INTO THE APPROPRIATE SINGULAR VALUES FOR EVALUATING STABILITY MARGINS, TOLERANCES TO STRUCTURED AND UNSTRUCTURED MODELING ERRORS, AND PERFORMANCE-ROBUSTNESS USING THE SO-CALLED STRUCTURED SINGULAR VALUE METHODOLOGY (MICROANALYSIS). FURTHERMORE, THROUGH THE USE OF THE SINGULAR VALUE DECOMPOSITION (SVD) THEOREM, THE EXPERIMENTAL DATA CAN BE USED TO DERIVE A COMPLETE MULTI-INPUT MULTI-OUTPUT DYNAMIC MODEL OF THE CLOSED-LOOP ADAPTIVE OPTICS SYSTEM.