【作者】 刘睿；

【导师】 王常虹；

【作者基本信息】 哈尔滨工业大学 ， 控制科学与工程， 2011， 博士

【摘要】 组合导航系统整合了不同的导航方式,实现优势互补,提高导航系统的精度和可靠性,已成为现代飞行器自主导航的重要手段,在军事和民用领域广泛应用。本文分析了惯性导航、地磁导航、天文导航的各自特点,对惯性/地磁/天文组合导航系统在飞行器自主导航领域中应用的理论和技术问题进行了深入研究。研究工作主要集中在以下几个方面:1)采用克里金插值方法构建出基准地磁图,该方法构建的地磁图具有较高的精度,可以准确反映构建区域内的地磁场信息,为进一步研究地磁导航技术提供了基础。通过分析相关算法在地磁场平缓区域应用时误匹配率高、实时性差等弊病的成因,提出了一种基于灰色关联的概率关联算法并分析了地磁基准图噪声和磁传感器测量噪声对定位误差的影响。该方法在地磁异常场微弱区域仍然具有良好的匹配定位性能,且不需要随时间累积或路径累积来进行匹配,提高了匹配的实时性,具有工程应用价值。2)针对地磁导航的特点,研究了在地磁异常场变化较剧烈区域基于EKF的惯性/地磁组合导航算法。分析了在地磁异常场被大大削弱地区采用拟合方法进行地磁导航时引起滤波发散的原因,提出了一种模糊自适应强跟踪卡尔曼滤波器,通过实时跟踪磁场量测噪声的变化,保证系统滤波精度,提高系统的稳定性。3)针对惯性导航系统的误差随时间累积的问题,提出了一种惯性/天文自主组合导航方法,该方法以天文高度角与方位角为观测量,利用惯性导航设备为天文导航提供水平基准,保证了算法的实时性,有效提高了导航系统的定位精度。针对工程应用中量测噪声统计特性可能改变的问题,提出了一种基于自适应交互多模型滤波算法。该算法无需预先确定模型集,当组合导航系统量测噪声统计特性变化时,能够以较少的模型数目实现对实际模态的覆盖,从而实现了对量测噪声统计特性变化的自适应调整。该算法较好地覆盖了实际模态变化情况,在观测值发生异常时,导航参数误差没有大的跳变,可以得到较高的滤波精度,并具有良好的稳定性,提高了整个导航系统的精度和可靠性。4)分析了全飞行时段惯性/天文/地磁组合导航算法的工作流程,并进行了仿真研究。针对惯性/天文/地磁组合导航系统因惯性/天文或惯性/地磁子系统发生故障导致整个系统被错误信息干扰的问题,采用一种容错滤波算法来提高导航系统的稳定性。针对惯导系统失效问题,提出了一种基于地磁/天文信息的飞行器自主导航方法。该方法利用地磁场的矢量特性,将星光与地磁矢量的夹角作为观测量,利用UKF滤波方法估计飞行器的导航参数。数值及半物理仿真研究表明:该方法是可行的,可用于平原,海洋等无明显地标特征的区域,也可以作为备份导航系统。更多还原

【Abstract】 Based on characteristics of multiple navigation modes, integrated navigation system can benefit from each of them, and can be more accurate and dependable. The integrated navigation system is the most important navigation mode at present. It is widely used in military and civilian area.Inertia, geomagnetism and celestial navigation systems are all analyzed respectively in this paper and this paper also analyze the theoretical and technical problem relate to realization of inertia/ geomagnetism/ celestial integrated navigation system. The research works focus on these points:1) Reference geomagnetic map is established based on high accuracy Kriging interpolation algorithm. Based on the analysis of the high mismatch rate and bad real-time when correlation algorithm is applied on gently area of geomagnetic field, a probability correlation algorithm rely on grey correlation is put forward, and the affect of positioning error comes from reference geomagnetic map and geomagnetic sensor noise is also analyzed. The algorithm has good performance even in anomaly magnetic field weak area, and the matching does not need the accumulation of time and path. The matching real-time is improved, and the algorithm is valuable in the application on engineering.2) Aim at the characteristic of geomagnetic navigation, an inertia/ geomagnetism integrated navigation algorithm based on EKF is studied. Analyze the causes that the filter is divergent when the interpolation algorithm is applied in the gently area of anomaly geomagnetic field. An adaptive fuzzy Kalman filter is put forward, it adjust the variance of measurement noise real-timely to keep the filter accurate and make the system stabile.3) Deal with the error accumulation of inertia navigation system, inertia/ celestial autonomous integrated navigation algorithm is put forward. This algorithm makes the celestial altitude and azimuth angle as the observed quantity, and the horizontal reference of which is provided by the inertia navigation equipment, so the algorithm is real-time and positioning accuracy of the navigation system is improved. Because the statistical characteristics of measurement noise may change when the algorithm is applied on engineering, a filter based on adaptive interacting multiple model is put forward. The application of this algorithm requires the model set should be pre-defined, and the actual modal should be covered by a model set as small as possible when the statistical characteristics of measurement noise changes, to ensure this algorithm is adaptive. It covers the change of actual modal well. Navigation parameter does not jump intensely when the observed quantities are abnormal, the filtering precision and stability are both high, and the whole navigation system is more accurate and dependable.4) Analyze the working process of full flight time inertia/ geomagnetism/ celestial integrated navigation algorithm, and test it by simulation. Based on the problem that the integrated navigation system is interrupt by error information when the inertia/ celestial or inertia/ geomagnetism subsystem is breakdown, a fault-tolerant filtering algorithm is put forward to improve the stability of the system. Deal with the breakdown of inertia navigation system, an autonomous navigation algorithm based on geomagnetism/ celestial information is put forward, this algorithm benefits form vectorialproperty of geomagnetism field, makes the angle between starlight and geomagnetism vector as its observed quantity, uses UKF to estimate the navigation parameter of the aircraft. Digital and the hardware-in-the-loop simulation both show that this algorithm is feasible. It can be used in either plain or ocean which has no obvious landmarks, and also can be the backup navigation system.更多还原