对先验地磁图的依赖是制约地磁导航发展的关键。受生物无需先验知识的地磁导航行为启发，研究地磁仿生导航问题。从仿生角度，将导航过程归结为多目标搜索问题，建立仿生导航模型。针对导航过程中，地磁场多参量与导航单一路径间存在的因果约束，提出了一种基于时序进化策略的搜索方法，使得多参量收敛伴随并指导导航搜索的进行。仿真结果证实了方法的有效性。

The a priori geomagnetic map is the key to the development of geomagnetic navigation.This paper re-searches for bio-inspired geomagnetic navigation inspired by animal navigation behavior without a priori knowledge. We generalize the bio-inspired geomagnetic navigation process as a multi-objective searching problem;then set up a bio-inspired navigation model.Considering causality constraint between the geomagnetic multi-parameters and the single navigation trajectory, we present a search mode using timing evolution searching strategy to make sure that the multi-parameters converge during search and direct navigation.The simulation results and their analysis confirm the validity of the method.

地磁/惯性组合导航为水下潜航器提供了一种高精度长航时自主定位的技术途径。当磁力仪安装在潜航器内部时,地磁测量的误差补偿及有效融合成为实现水下地磁/惯性组合导航的关键技术。文中提出了基于模标定的地磁测量误差补偿技术,实现磁力仪标定并削弱潜航器磁场的干扰。设计点群滤波和扩展卡尔曼滤波两级滤波算法,实现了大观测噪声地磁测量的有效融合。水下试验表明,地磁测量误差补偿技术能够将干扰磁场从10000 nT降低至100 nT,采用两级滤波算法的组合导航系统能够取得300 m的定位精度。

Geomagnetic field/inertial integrated navigation provides a technique way for high precision autonomous localization of underwater vehicles with long duration voyage. In the case that the magnetometer is installed inside of the underwater vehicles, the compensation of geomagnetic measuring errors and the fusion of compensated geomagnetic measurements with the inertial information are essential to the integrated navigation. In this paper, the compensation technique based on scalar calibration was proposed to calibrate the magnetometers and reduce the effects of the vehicle magnetic field. Combined point mass filter and extended Kalman filter were designed to fuse the geomagnetic measurements with large noises. The results of the underwater experiment show that the disturbing magnetic magnitude is reduced from ten thousand nanoTeslas to several tens of nanoTeslas, and the localization accuracy of 300 m is achieved with the combined filtering algorithm.

惯性导航技术是一种完全自主的导航技术，为了解决惯性导航定位误差随时间积累不断增加的问题，采取地磁导航的方式来修正惯性导航的误差。研究地磁信息的熵算法和ICCP算法，利用熵算法具有剔除离散点、野值点，并且抗基准误差能力强和ICCP算法具有匹配精度较高，匹配结果较稳定的优点，提出一种将熵算法和改进的ICCP算法融合的新算法，以实现粗精结合多级匹配。仿真结果表明，该算法能够取得了很好的匹配效果，在一定程度上修正惯导误差，抑制惯性导航定位误差的发散，提高整个导航的精度，实现惯性导航长时间工作。

Inertial navigation technology is a completely autonomous navigation technology.But the positioning errors of the inertial navigation accumulate when time increasing.In order to solve the problem,geomagnetic is taken to correct the inertial navigation errors in navigation.This paper studies the entropy and ICCP algo-rithm,and then proposes a multi-level geomagnetic match algorithm based on entropy/ICCP algorithm.This algorithm uses a way of rough and precise combination with multi-level multi-scale matching.The Entropy algorithm has the ability of removed the discrete points,outliers,and strong ability of anti-reference error. ICCP algorithm has the characteristics of high precision and stable match results.The simulation experiment successfully proved the algorithm can get good match results.It improves the accuracy of the entire navigation and achieves inertial navigation’s long hours working by correcting and suppressing the divergence positioning errors of INS in some ex

区域地磁变化场建模与预测是地磁辅助导航、空间环境监测与预报等领域的重要研究课题。结合时域上的单台站地磁变化场预测,以及空间域上的区域地磁变化场重构,提出了一种基于克里格重构的区域地磁变化场短时预测模型。以区域范围(101°E~111°E,30°N~40°N)的地磁测站2009年的数据为实例,结果表明:1)基于克里格方法重构的区域地磁变化场Z分量,准确合理的反映了其时空变化特性;2)在短时预测中,区域模型能紧跟地磁变化场的变化趋势,说明其能很好地描述变化磁场的时空演化特性,且位于区域边缘的台站预测误差稍大。当预测期为1 h、6 h、12 h、24 h时,区域地磁变化场预测模型的平均绝对误差分别为1.67 nT、2.19 nT、2.72 nT、3.14 nT。

The modeling and forecasting of regional geomagnetic variation field is an important research topic of geomagnetic navigation and space environment monitoring. Combining the time-domain forecasting of the single geomagnetic observatory and the space-domain reconstruction of the regional geomagnetic variation field, this paper proposes a regional geomagnetic forecasting model based on Kriging reconstruction method. It shows that: 1) the regional geomagnetic Z component reconstructed by Kriging method can reasonably and accurately reflect its spatial-temporal variation characteristics; 2) the regional model can closely keep up with the trend of geomagnetic variation field in short-term forecasting, which indicates that the model can accurately describe the spatial-temporal evolution characteristics. The forecasting errors of geomagnetic observatory located in the area edge are larger. When the forecasting period is 1 h, 6 h, 12 h, 24 h, the mean absolute error of the regional model is 1.

基于亥姆霍兹线圈产生均匀磁场的特性,绕制了三轴亥姆霍兹线圈,通过改变3组线圈通电电流大小,分别产生3个正交方向的磁场,模拟地磁场强度,为地磁导航系统的半实物仿真实验提供地磁场环境。同时,就地磁导航半实物仿真的需求,对地磁场模拟生成装置进行了可重复性和线性度验证实验、响应速度测量实验、均匀性验证实验和稳定性验证实验,并建立了装置的输入输出模型,验证了装置的各项指标符合实验室需求。

Based on the feature that Helmholtz coils can produce a uniform magnetic field,we made triaxial Helmholtz coils to generate magnetic fields with three orthogonal directions for simulating the geomagnetic field intensity by changing the importing electrical current of the three coils .This device could offer geomagnetic field environment for the hardware-in-the-loop simulation of geomagnetic navigation system .A lot of experiments were made to find whether this device is suitable for the need of simulation .These experiments included the tests on repeatability and linearity,and on response speed,uniformity and stability .The input/output model of the device was also established .These experiments proved that the indexes of the device can satisfy the needs of the laboratory .