摘要
为缩小惯导初始指示航迹与真实航迹偏差,有效提高水下重力匹配导航精度,降低失误风险,本文将粒子滤波与粒子群优化算法相结合,分析了粒子滤波算法中不同蒙特卡罗采样样本量、不同重力异常观测误差、不同惯导指示航迹初始偏差对重力匹配导航结果精度的影响。联合粒子群优化改进算法,削弱了初始配准误差对后续粒子滤波的影响。将初始航迹匹配误差作为平移参数,修正惯性导航整体航迹,提升了粒子滤波算法的匹配导航精度。本文通过算例分析,进一步验证了联合算法的可行性和有效性。
To reduce the deviation between the initially indicated track of the inertial navigation and the real track, effectively improve the underwater gravity matching navigation accuracy, and reduce the risk of errors, the particle filtering algorithm and the particle swarm optimization algorithm are combined in this paper to analyze the influence of different Monte Carlo sample sizes, different gravity anomaly observation errors, and initial deviations of different inertial navigation indicated tracks in the particle filtering algorithm on the accuracy of gravity matching navigation results. The particle swarm optimization algorithm is adopted to reduce the initial registration error impact on the subsequent particle filtering. The initial track matching error is taken as the translation parameter to correct the whole track of inertial navigation and improve the gravity matching navigation accuracy of the particle filtering algorithm. The feasibility and effectiveness of the joint algorithm are further verified through an analysis of examples.
作者
欧阳明达
杨元喜
OUYANG Mingda;YANG Yuanxi(Institute of Geospatial Information,Information Engineering University,Zhengzhou 450052,China;Stake Key Laboratory of Geo-information Engineering,Xi′an 710054,China)
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
北大核心
2022年第10期1514-1521,共8页
Journal of Harbin Engineering University
基金
国家自然科学基金重点项目(41931076)
国家自然科学基金项目(42174001)
地理信息工程国家重点实验室自主研究课题项目(SKLGIE2020-ZZ-4)。
关键词
线性滤波
粒子滤波
粒子群优化算法
重力匹配导航
蒙特卡罗采样
贝叶斯估计
惯性导航系统
linear filtering
particle filtering
particle swarm optimization algorithm
gravity matching navigation
Monte Carlo sampling
Bayesian estimation
inertial navigation system
