多传感器组合导航系统信息融合方法研究

发布时间：2018-07-17 15:32

【摘要】：利用卫星导航系统和捷联惯性导航系统多传感器进行组合导航达到高精度导航要求在民用领域和军用领域是国内外研究的热点。与此同时,信息融合方法的突飞猛进为提升组合导航的精度提供了重要支撑。本文基于北斗卫星导航系统和捷联惯性导航系统组合导航的紧组合方式开展信息融合方法研究。信息融合方法主要将北斗卫星导航系统的伪距和伪距率与捷联惯性导航系统的导航数据进行时空配准,数据关联和数据融合。本文在数据关联方面,提出了一种自适应概率数据关联法。自适应概率数据关联法提出适应导航组合新息的自适应加权的修正参数,并通过仿真实验分析可得出自适应概率数据关联法的数据关联效果要优于其他关联算法。自适应概率数据关联法选择不同的参数,其数据关联效果也有所不同,因此在实际应用中要根据实际情况选择合适的参数来达到最佳的数据关联效果。本文在数据融合方面,在粒子群优化粒子滤波的基础上运用一种修正权重粒子群优化粒子滤波算法通过加入优势速度和劣势速度来优化粒子的更新模式,达到修正改变粒子的权重,来综合粒子群优化粒子滤波算法的全局和局部的搜索能力,让粒子收敛加快,减小粒子滤波中陷入局部最优的概率,可以在较短的时间内实现全局精确定位。本文最后对北斗卫星导航系统和捷联惯性导航系统组合导航的紧组合方式信息融合进行仿真实验对比,实验表明经过数据关联和数据融合两个重要步骤的优化,改进算法的性能表现无论是在空间位置上,还是空间速度上的误差估计和精度上都表现出明显的优势。
[Abstract]:Multi-sensor integrated navigation using satellite navigation system and strapdown inertial navigation system to achieve high accuracy navigation requirements in civil and military fields is a research hotspot at home and abroad. At the same time, the rapid development of information fusion method provides an important support for improving the accuracy of integrated navigation. In this paper, the information fusion method is studied based on the tight combination of Beidou Satellite Navigation system and Strapdown Inertial Navigation system. The information fusion method mainly uses the pseudo-range and pseudo-range rate of Beidou satellite navigation system and the navigation data of strapdown inertial navigation system for space-time registration, data association and data fusion. In this paper, an adaptive probabilistic data association method is proposed for data association. Adaptive probabilistic data association method proposed adaptive weighted parameters for navigation integrated innovation, and through simulation analysis, it can be concluded that the data association effect of adaptive probabilistic data association method is better than that of other association algorithms. The adaptive probabilistic data association method selects different parameters, and the data association effect is different. Therefore, the optimal data association effect should be achieved by selecting the appropriate parameters according to the actual situation in the practical application. In this paper, in the aspect of data fusion, a modified weighted particle swarm optimization particle filter algorithm is used to optimize the particle updating mode by adding the superior speed and inferior speed on the basis of particle swarm optimization particle filter. By modifying and changing particle weight to synthesize the global and local searching ability of particle swarm optimization particle filter algorithm, the convergence of particles can be accelerated and the probability of falling into local optimum in particle filter can be reduced. Global positioning can be achieved in a short time. In the end of this paper, the information fusion of the tight integrated mode of Beidou satellite navigation system and strapdown inertial navigation system is simulated and compared. The experiment shows that the data association and data fusion are optimized by two important steps: data association and data fusion. The performance of the improved algorithm shows obvious advantages both in the spatial position, in the error estimation and accuracy of the spatial velocity.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN967.2

【相似文献】