Optimization on Attitude Algorithms for Filtered Data in Dither RLG SINS

In the dither ring laser gyro (DRLG) strapdown inertial navigation system (SINS), gyro outputs are often pre-filtered to eliminate the dither modulated, but the influence of the filtered data on the attitude algorithm Is often ignored. It is found that the conventional coning algorithms can not work to high accuracy for the filtered gyro outputs which have a shaped frequency. So the coning algorithm must be optimized to match the frequency response of the dither demodulation filter. The optimized attitude algorithm formulas that can be applied for the filtered gyro data are derived. It has the same form as the conventional algorithm and only the coefficients of incremental angle vector cross products are amended. The algorithm compensation effect is relative to the characteristics of the dither demodulation filter. The coefficients of optimized algorithm are obtained for a FIR filter which is used to demodulate the gyro dither. The error characteristics of conventional and optimized algorithms are compared and analyzed. The ideal simulation and real experiments illustrate that the optimized algorithms accuracy is remarkably improved for filtered gyro data and the algorithm have great value to real application in DRLG SINS.