一种基于Haar小波的光纤陀螺信号实时滤波方法

Aiming at improving the quality of fiber optic gyroscope (FOG) signal in the inertial stabilization platform control system, a real-time software filtering method of FOG signal based on Haar wavelet is proposed. Firstly, it is illustratedby sufficient simulations that the boundary problem isthe source that affects the real-time filtering effect of existing wavelet methods. Then, through theoretical analysis, the reason for the boundary problem is analyzed, and Haar wavelet's feasibility for real-time filtering of FOG signal is certified from two aspects, i. e. the support set and the continuous ladder signal approximation. Finally, the real-time gyro signal filtering method based on Haar wavelet is proposed, which not only takes advantage of the strong signal-noise separation capability of wavelet, but also can meet the requirements of real-time application background, and is fully applicable to the signal filtering of FOG in the control system of the inertial stable platform. The experiment shows that the proposed method has good real-time filtering performance, and the random noise of the filtered signal can be reduced by more than 40%. After using filtered gyro signal in closed loop control, the stability accuracy of the inertial stabilization platform control system can be increased by 14%.