Self-Adapative Notch Filter for the V-22 Flight Controls using Stochastic Realization Algorithm (SRA)

This proposal addresses the problem of mitigating the effect of unwanted structural modes of the V-22 aircraft on the performance of the flight control system. The current system designed by Bell Textron uses five nonself-adaptive filters in the frequency range of 2.3-5.5Hz on SAHRS and VYROS outputs and pilot stick control inputs to remove the effect of unwanted structural modes. The filter frequencies are scheduled with airspeed and conversion angle but not with respect to gross weight. The purpose of designing a self-adaptive notch filter is to compensate for changes in frequency due to effects such as gross weight and to achieve narrow filters with minimal phase lag. The use of a System Identification approach based on the Stochastic Realization Algorithm (SRA) is proposed. SRA has been shown to be very effective in estimating the parameters (frequencies, dampings, mode shapes) of structural modes of aircraft and flexible structures using measurement of outputs only. SRA does not require any special inputs and produces accurate results even with short data lengths and under poor signal-to-noise ratios (SNR) conditions. SRA has been tested on NASA flight test data from X-29 and F-18 and found to outperform other methods of structural mode parameter identification. The design of a self-adaptive notch filter using SRA results is based on a Kalman Filter. During Phase I, the feasibility of removing the unwanted effects of the structural modes using SRA-KF will be demonstrated on simulated and flight test data from the V-22 aircraft. Phase II will involve hardware/software integration and flight testing. Bell Textron, designer of structural filters for the V-22 flight control system is expected to support the proposed effort. (Letter enclosed)