基于自适应FLP滤波的金属谐振陀螺信号处理方法  

Signal processing method of metal resonant gyro based on adaptive FLP filter

作　　者：刘为任[1] 张海龙 侯志宁[1] 王宁[1] 张浩 LIU Weiren;ZHANG Hailong;HOU Zhining;WANG Ning;ZHANG Hao(Tianjin Navigation Instrument Research Institute,Tianjin 300131,China)

机构地区：[1]天津航海仪器研究所,天津300131

出　　处：《中国惯性技术学报》2019年第1期83-88,共6页Journal of Chinese Inertial Technology

基　　金：天津市工业企业专项基金(201707129)

摘　　要：针对传统前向线性预测(FLP)滤波在金属谐振陀螺信号处理中,无法解决收敛速度和稳态误差之间矛盾且动态滤波效果较差的问题,提出了一种基于自适应FLP滤波信号处理方法。首先采用Allan方差对陀螺原始信号进行分析,得出不同噪声占比;然后采用自适应FLP滤波方法,通过建立补偿因子与估计误差的非线性关系,实时调整步长因子大小,修正滤波误差;最后用自研的金属谐振陀螺惯性系统试验验证该方法的有效性。试验结果表明:自适应FLP滤波方法可抑制量化噪声、角度随机游走和零偏不稳定性等。降噪效果比传统FLP滤波方法更为明显,收敛速度更快,动态时自适应FLP滤波方法降噪效果是传统FLP滤波方法的1.95倍,该方法可有效提高金属谐振陀螺的性能。In view that traditional forward linear prediction(FLP) filtering can not solve the contradiction between the convergence speed and the steady state error, and has poor dynamic filtering effect, a signal processing method based on adaptive FLP filter is proposed. Firstly, Allan variance is used to analyze the gyro’s original signal, and different noise ratios are obtained. Then, the adaptive FLP filtering method is adopted to adjust the step size factor in real time and correct the filtering error by establishing the nonlinear relationship between the compensation factor and the estimation error. Finally, the validity of the proposed method is verified by a self-developed metal resonant gyro inertial system test. Experimental results show that the proposed method can suppress quantization noise, angle random walk and zero bias instability, etc.The noise reduction effect is more significant than that of the traditional FLP filtering method. The denoising effect by the proposed method in dynamic state is 1.95 times of that by the traditional FLP filtering method. The proposed method can effectively improve the performance of metal resonant gyro with faster convergence speed.

关 键 词：金属谐振陀螺 自适应滤波 前向线性预测滤波 信号处理 

分 类 号：U666.1[交通运输工程—船舶及航道工程]