初始对准是惯性导航领域的重要研究课题，由于MEMS陀螺精度相对较低，噪声较大，地球自转角速率湮没在噪声之中，因此无法用传统的高精度捷联式惯性导航系统（SINS）的双矢量定姿的方法进行对准。现阶段，大多采用磁罗经、卫星天线等提供航向，但是都存在容易受到环境干扰的不足。旋转调制技术的引入使得MIMU的自对准成为可能。本文重点开展了在静基座及晃动基座条件下MIMU的自对准算法、旋转状态下的可观测性研究、小型一体化的MEMS陀螺寻北仪的实现。主要研究内容和研究成果如下：针对本文重要基础的旋转调制技术开展了研究，对其误差抑制原理作了较为全面地研究。在旋转方案分析基础上，对其误差抑制效果进行了分析，并对影响旋转调制技术误差抑制效果的一些因素进行了讨论。针对MIMU特点及应用场景，提出基于Allan方差的最优加权融合算法，解决融合算法中确定陀螺权重的问题，提高了航向精度；提出了多个MIMU在连续旋转方案下的航向精度计算公式并进行了验证，解决了航向精度指标确定的情况下，如何确定陀螺精度水平、MIMU配置与测量时长这三者之间的关系。针对目前MIMU的可观测性研究不够完善这一不足，分析了多种旋转方案下其状态可观测性情况。通过奇异值分解（SVD）定量分析了一些可观测程度弱的状态量在MIMU旋转情况下的可观测度变化情况，为旋转方案提供了理论指导，并通过仿真分析和实测数据进行了验证。提出了一种MIMU抗晃动干扰的自对准方法，通过基于旋转调制的惯性空间重力矢量对准（RMT-ISGBA）算法架构，解决了传统寻北算法易受外界干扰的不足。研究了惯性器件的常值偏差、陀螺的角度随机游走这两项主要误差因素对算法精度的影响。在室外车载干扰环境下航向达到优于1°的精度，验证了算法的鲁棒性；进一步地引入互补滤波算法提高了水平姿态角的精度，使得水平方向达到优于0.05°的精度，验证了方法的有效性。针对目前MEMS陀螺寻北仪体积较大、功耗较高的不足，提出了合理的小型一体化的结构设计方案，采用PCB盘式电机、无线传输、电容式角位移传感器等方案减小体积，实现了小体积低功耗的旋转调制MEMS陀螺寻北仪原理样机。总功耗小于2W，体积为112.8 cm3，3分钟能够达到优于0.5°的航向精度。

Initial alignment is an important research topic in the field of inertial navigation. Due to the relatively low accuracy and large noise of MEMS gyros, the measured earth's rotation rate is immersed in the noise, so traditional double-vector attitude determination of high end strapdown inertial navigation systems (SINS) for alignment cannot be used. At present, magnetic compasses and satellite antenna are always used to provide heading. However, these are vulnerable to environmental interference.The introduction of rotation modulation technology (RMT) makes MIMU self-alignment possible. This paper focuses on the self-alignment algorithm of MIMU under the condition of static base and swing base, the observability analysis under rotating state, and the realization of small integrated MEMS gyro north finder. The main research contents are as follows:The research on the RMT is carried out, and its error suppression is studied. Based on the analysis of the rotation scheme, its error suppression effect is verified, and some factors affecting the error suppression effect of the RMT are analyzed.According to the characteristics and application scenarios of MIMU, an optimal weighted fusion algorithm based on Allan variance is proposed to solve the problem of determining the gyro weight in the fusion algorithm and improve the accuracy of heading. The calculation formula of heading accuracy of multiple MIMUs under the continuous rotation scheme is proposed and verified, which solves the relationship among gyro accuracy, configuration of MIMUs and measurement time when the heading accuracy index is determined.In view of the lack of observability research of MIMU, the state observability based on multiple rotation schemes is analyzed. The singular value decomposition (SVD) was used to quantitatively analyze the observability of some weakly observable state quantities when the MIMU was rotating, which provided theoretical guidance for the rotation scheme and was verified by simulation analysis and measured data.The inertial space gravity based alignment using rotation modulation technology (RMT-ISGBA) algorithm with anti-interference ability is adopted, which solves the shortcomings of the north finding algorithm that is susceptible to external interference. The constant bias and angle random walk (ARW) of the gyro which are the two main error factors on the accuracy of the algorithm are studied. It was analyzed and verified in the outdoor vehicle environment. The accuracy of the heading is better than 1 °, which verifies the robustness of the algorithm. The quaternion based complementary filtering (QBCF) algorithm is introduced to further improve the accuracy of the horizontal attitude angles, and the accuracy is better than 0.05 °, which verifies the effectiveness of the method.Aiming at the shortcomings of the current MEMS gyro north-finder prototype with large volume and high power consumption, a reasonable small-scale integrated structural design scheme is proposed, which uses PCB disk motor, wireless transmission, capacitive angular sensor and other technologies to reduce the volume. A small volume and low power consumption rotary modulation MEMS gyro north finder is realized. The total power consumption is less than 2W, the volume is 112.8 cm3, and the heading accuracy is better than 0.5 °in 3 minutes.