水下运载体重力匹配算法研究

发布时间：2017-12-28 21:14

本文关键词：水下运载体重力匹配算法研究　出处：《北京理工大学》2016年硕士论文　论文类型：学位论文

【摘要】：重力匹配算法是重力辅助惯性导航系统的核心,它将实测重力异常数据与惯导输出信息按照一定方法进行相关性分析,从而估计载体位置,校正惯导误差。传统的序列匹配算法实时性不高,单点匹配算法在重力异常变化明显的区域容易发散。并且传统匹配算法相邻匹配点之间相互独立,出现误匹配时不能及时修正,因此提高匹配精度和保证匹配结果的可信度对长航时惯导系统的误差校正有着重要意义。本文基于单点粒子滤波算法,将惯导相邻点之间的位置相关性加入到匹配过程中,克服了传统匹配算法易出现误匹配的缺点,算法的精度更高鲁棒性更强。主要工作和创新点如下:(1)提出在单点匹配算法中利用粒子滤波代替扩展卡尔曼滤波。传统的单点匹配算法桑迪亚算法所利用的扩展卡尔曼滤波由于线性化误差,在重力异常变化大的区域易发散。针对此问题,采用粒子滤波代替扩展卡尔曼滤波避免线性化带来的误差,扩大了单点匹配算法的适用范围。仿真实验表明该方法能够避免传统桑迪亚匹配算法在重力异常变化大的区域易发散的缺点,提高匹配精度。(2)基于单点粒子滤波算法,提出了定尺度定方向的矢量匹配算法,该方法可以避免传统匹配方法中出现的误匹配现象。在单点粒子滤波匹配结果的基础上加入惯导相邻点之间的距离相关性,每个匹配点的匹配结果由在匹配点周围所选择的若干小于匹配时刻的采样点的匹配结果进行校正,借鉴贪心算法,为减小随机误差的影响,利用加权最小二乘法求得最终的匹配结果。仿真实验表明,该算法与传统匹配算法相比,具有更高的精度和更强的鲁棒性。(3)在定尺度定方向矢量匹配算法基础上,为扩大算法的适用范围和提高精度,提出了变尺度变方向矢量匹配算法。为适应水下载体的机动性,改进可信度判别式使得算法适用于水下载体的非匀速直线航行。并为提高算法在载体转弯处的精度,将大于匹配时刻的采样点加入到对当前匹配结果的校正过程中。并在此基础上加入惯导相邻点之间的相位相关性,进一步提高算法精度。仿真实验表明,该算法在惯导航迹为非匀速直线运动时的精度和鲁棒性都明显提高。
[Abstract]:Gravity matching algorithm is the core of gravity aided inertial navigation system. It performs correlation analysis between measured gravity data and inertial navigation output information in a certain way, so as to estimate position of carrier and correct inertial navigation error. The real-time performance of the traditional sequence matching algorithm is not high, and the single point matching algorithm is easy to diverge in the area with the obvious variation of gravity anomaly. Moreover, the matching points between traditional matching algorithms are independent of each other, which can not be corrected in time. Therefore, improving the matching accuracy and ensuring the confidence of matching results is of great significance for the error correction of long haul inertial navigation system. Based on the single point particle filter algorithm, the location correlation between adjacent points of inertial navigation is added to the matching process, which overcomes the shortcoming of the traditional matching algorithm which is prone to mismatch. The algorithm is more precise and robust. The main work and innovation are as follows: (1) it is proposed to use particle filter instead of extended Calman filter in the single point matching algorithm. The extended Calman filter, which is used by the traditional single point matching algorithm Sandy subalgorithm, is easy to diverge in the region with large gravity anomaly because of the linearization error. In order to solve this problem, the particle filter is used instead of extended Calman filter to avoid the error caused by linearization, and the application scope of the single point matching algorithm is expanded. The simulation experiment shows that the method can avoid the disadvantages of the traditional Sandy submatching algorithm in the area with large gravity anomaly, and improve the matching precision. (2) based on the single point particle filter algorithm, a fixed scale and direction vector matching algorithm is proposed, which can avoid the mismatch phenomenon in the traditional matching method. Based on the matching results of single point particle filter with inertial distance correlation between adjacent points, each sampling point matching results point by the selected point in the match around some time, less than the matching result is corrected using the greedy algorithm, in order to reduce the influence of random error, the final matching result by using the weighted least squares method. The simulation experiment shows that the algorithm has higher accuracy and stronger robustness than the traditional matching algorithm. (3) on the basis of the fixed scale fixed direction vector matching algorithm, a variable scale variable direction vector matching algorithm is proposed in order to enlarge the application range and improve the precision of the algorithm. In order to adapt to the mobility of the water downloading body, the improved reliability discriminant makes the algorithm applicable to the non uniform linear navigation of the water downloading body. In order to improve the accuracy of the algorithm at the turning point of the carrier, the sampling point which is larger than the matching time is added to the correction process of the current matching result. On this basis, the phase correlation between the adjacent points of inertial navigation is added to further improve the accuracy of the algorithm. The simulation results show that the accuracy and robustness of the algorithm are greatly improved when the inertial navigation track is moving in a non uniform linear motion.
【学位授予单位】：北京理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U666.1

【相似文献】