基于滑动窗迭代最大后验估计的多源组合导航因子图融合算法

doi:10.3969/j.issn.1000-1093.2019.04.016

摘要/Abstract

摘要： 在应用因子图算法完成多源组合导航数据融合的过程中，子系统观测噪声的时变特性将对导航状态估计的准确性产生极大影响。为解决这一问题，提出一种基于高斯模型下的子系统观测量均值向量和协方差矩阵的估计算法。该算法利用因子图最优化过程中每个迭代周期下的观测量残差，实时地更新各个子系统观测量的均值向量和协方差矩阵的最大后验估计值，从而得到更加准确的导航状态估计值，在提出新算法的同时也验证了新算法对最优化过程收敛性的影响。仿真测试与实验测试结果表明，与已有的标准因子图算法、基于最大似然估计的因子图算法和基于最大后验估计的因子图算法相比，所提出的基于迭代最大后验估计的因子图算法能够有效提高子系统观测状态变化时的多源组合导航估计精度。

关键词: 多源组合导航, 信息融合, 因子图, 惯性导航系统, 全球卫星导航系统, 超宽带导航系统

Abstract: In the process of data fusion in multi-source integrated navigation using factor graph, the time-varying characteristics of the subsystem's observed noise have a great influence on the estimation accuracy of navigation state. In order to solve the problem, a Gaussian model-based method to estimate the mean vector and covariance matrix of sub-system observation is proposed. In the proposed method, the observed-measurement residuals for each iterative cycle in the process of factor graph optimization are utilized to update the maximum posteriori estimated values of mean vectors and covariance matrices. A more accurate estimated value of navigation state can be obtained by estimating the sub-system noise state. The influence of the new algorithm on the convergence of optimization process was also deduced. Both the simulated and experimental results show that, compared with the existing algorithms as factor graph, maximum likelihood estimation based factor graph and maximum posteriori based factor graph, the proposed factor graph method based on iterative maximum posteriori estimation can effectively improve the accuracy of navigation estimation when the subsystem observing state varies. Key

Key words: multi-sourceintegratednavigation, datafusion, factorgraph, inertialnavigationsystem, globalnavigationsatellitesystem, ultra-widebandnavigationsystem

中图分类号:

V249.32⁺8

徐昊玮，廉保旺，刘尚波. 基于滑动窗迭代最大后验估计的多源组合导航因子图融合算法[J]. 兵工学报, 2019, 40(4): 807-819.

XU Haowei, LIAN Baowang, LIU Shangbo. Multi-source Integrated Navigation Algorithm for Iterated Maximum Posteriori Estimation Based on Sliding-window Factor Graph[J]. Acta Armamentarii, 2019, 40(4): 807-819.

参考文献

［1］ LI Z, WANG R, GAO J. An approach to improve the positioning performance of GPS/INS/UWB integrated system with two-step filter[J]. Remote Sensing, 2017, 10(1): 19-34.
[2] WANG M, LI Y, FENG G, et al. Key technologies of GNSS/INS/VO deep integration for UGV navigation in urban canyon[C]∥Proceedings of the 11th Asian Control Conference. Gold Coast, Australia: IEEE, 2017: 2546-2551.
[3] XU H W, LIAN B W, TOTH C K, et al. An airborne LiDAR/INS integrated navigation algorithm based on fuzzy controlled SIFT[C]∥Proceedings of the 2017 International Technical Meeting of the Institute of Navigation. Monterey. CA, US: the Institute of Navigation, 2017: 313-326.
[4] YUAN G M, YUAN K F, ZHANG H W. A variable proportion adaptive federal Kalman filter for INS/ESGM/GPS/DVL integrated navigation system[C]∥Proceedings of the 4th International Joint Conference on Computational Sciences and Optimization. Kunming/Lijiang, Yunnan, China: IEEE, 2011: 978-981.
[5] 钱进.改进的联合Kalman滤波及其在组合导航中的应用[J].火力与指挥控制,2009,34(7):142-144.
QIAN J. A modified federal Kalman filter algorithm and it's application at integrated navigation system[J]. Fire Control and Command Control, 2009, 34(7):142-144. (in Chinese)
[6] XU H W, LIAN B W. Fault detection for multi-source integrated navigation system using fully convolutional neural network[J]. IET Radar, Sonar & Navigation, 2018, 12(7):774-782.
[7] GAO W, ZHANG Y, WANG J. A strapdown inertial navigation system/Beidou/Doppler velocity log integrated navigation algorithm based on a cubature Kalman filter[J]. Sensors, 2014, 14(1): 1511-1527.
[8] 何伟, 廉保旺, 冯晓明. 基于联邦卡尔曼滤波的组合导航定位算法[J]. 火力与指挥控制, 2012, 37(8):147-150.
HE W, LIAN B W, FENG X M. Integrated navigation and location algorithm based on Federated Kalman filter[J]. Fire Control and Command Control, 2012, 37(8):147-150. (in Chinese)
[9] 高军强, 汤霞清, 张环,等. 基于因子图的车载INS/GNSS/OD组合导航算法[J]. 系统工程与电子技术, 2018, 40(11): 2547-2554.
GAO J Q, TANG X Q, ZHANG H, et al. Vehicle INS/GNSS/OD integrated navigation algorithm based on factor graph[J]. Systems Engineering and Electronics, 2018, 40(11): 2547-2554.(in Chinese)

[10] INDELMAN V, WILLIAMS S, KAESS M, et al. Information fusion in navigation systems via factor graph based incremental smoothing[J]. Robotics and Autonomous Systems, 2013, 61(8): 721-738.
[11] ZENG Q, CHEN W, LIU J, et al. An improved multi-sensor fusion navigation algorithm based on the factor graph[J]. Sensors, 2017, 17(3): 641-656.
[12] 王慧哲, 曾庆化, 刘建业, 等. 基于因子图的无人机全源导航关键技术研究[J]. 导航与控制, 2017, 16(2):1-5.
WANG H Z, ZENG Q H, LIU J Y, et al. Research on the key technology of UAV of all source position navigation based on factor graph[J]. Navigation and Control, 2017, 16(2):1-5. (in Chinese)
[13] 吕瑞, 陈龙, 翁雪, 等. 利用先验点图模型的SLAM后端优化算法[J]. 武汉大学学报(信息科学版), 2014, 39(6):745-749.
L R, CHEN L, WENG X. A back-end optimization algorithm of SLAM based on graph model with prior points[J]. Geomatics and Information Science of Wuhan University, 2014, 39(6):745-749. (in Chinese)
[14] CHIU H, WILLIAMS S, DELLAERT F, et al. Robust vision-aided navigation using sliding-window factor graphs[C]∥Proceedings of 2013 IEEE International Conference on Robotics and Automation. Karlsruhe, Germany: IEEE, 2013: 46-53.
[15] CROUCH M S, MCGREGOR A, STUBBS D. Dynamic graphs in the sliding-window model[M]. London, UK: Springer, 2013.
[16] 秦永元, 张洪钺，汪叔华. 卡尔曼滤波与组合导航原理[M]. 西安：西北工业大学出版社, 1998:277-281.
QIN Y Y, ZHANG H Y, WANG S H. Theory of Kalman filter and integrated navigation[M].Xi'an: Northwestern Polytechnic University Press, 1998:277-281.(in Chinese)
[17] INDELMAN V, WILLIAMS S, KAESS M. Factor graph based incremental smoothing in inertial navigation systems[C]∥Proceedings of the 15th International Conference on Information Fusion. Singapore: IEEE, 2012: 2154-2161.
[18] PAN J X, FANG K T. Growth curve models and statistical diagnostics[M]. New York, NY, US: Springer, 2002.
[19] ELISHA Y B, INDELMAN V. Active online visual-inertial navigation and sensor calibration via belief space planning and factor graph based incremental smoothing[C]∥Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Vancouver, Canada: IEEE, 2017: 2616-2622.
[20] LIU H Y, ZHANG Z Y, GRIMM K J. Comparison of inverse Wishart and separation-strategy priors for Bayesian estimation of covariance parameter matrix in growth curve analysis[J]. Structural Equation Modeling: A Multidisciplinary Journal, 2016, 23(3): 354-367.
[21] HARTLEY H O. The modified Gauss-Newton method for the fitting of non-linear regression functions by least squares[J]. Technometrics, 1961, 3(2): 269-280.
[22] ECKENHOFF K, GENEVA P, HUANG G. Direct visual-inertial navigation with analytical preintegration[C]∥Proceedings of IEEE International Conference on Robotics and Automation. Marina Bay Sands, Singapore: IEEE, 2017: 1429-1435.

第40卷
第4期2019 年4月兵工学报ACTA
ARMAMENTARIIVol.40No.4Apr.2019

[1]	周可歆，丁旭辉，吕德东，卜祥元，安建平. 天基信息传输高效低复杂度LDPC译码算法研究[J]. 兵工学报, 2024, 45(4): 1176-1185.
[2]	王博，蒋平，郭波. 基于多源信息融合的航天阀门可靠性评估[J]. 兵工学报, 2022, 43(1): 199-206.
[3]	张羽，王朋，刘纪元，钟荣兴，韦琳哲，迟骋. 合成孔径声纳重叠相位中心与惯性导航系统联合估计运动误差算法[J]. 兵工学报, 2021, 42(3): 588-597.
[4]	戴天，缪玲娟，邵海俊. 一种基于多波束测深仪的水下同步定位与地图构建方法[J]. 兵工学报, 2020, 41(7): 1384-1392.
[5]	沈世斌，谢非，赵静，钱伟行，康国华，刘锡祥. 基于相位控制的惯性与卫星超紧组合导航系统信号解调方法[J]. 兵工学报, 2020, 41(3): 495-506.
[6]	胡杰，严勇杰，王子卉. 基于速度约束与模糊自适应滤波的车载组合导航[J]. 兵工学报, 2020, 41(2): 231-238.
[7]	王琪，汪立新，梁树晖，秦伟伟，沈强. 基于可观测性分析的混合式惯性导航系统连续自标定模型选择[J]. 兵工学报, 2020, 41(1): 56-67.
[8]	李洪瑞. 基于神经网络的分布式被动传感器信息融合技术[J]. 兵工学报, 2020, 41(1): 95-101.
[9]	朱兵，许江宁，吴苗，李京书，何泓洋，李峰. 基于逆向导航算法的捷联式惯性导航系统改进优化对准方法[J]. 兵工学报, 2019, 40(9): 1860-1870.
[10]	董春梅，任顺清，陈希军，李巍. 一种激光陀螺捷联惯性导航系统级标定方法[J]. 兵工学报, 2019, 40(8): 1618-1626.
[11]	闫理跃, 王厚军, 刘震. 一种结合噪声辅助技术和现场数据的模拟电路实时可靠度预测方法[J]. 兵工学报, 2019, 40(2): 326-333.
[12]	李群生，赵剡，王进达. 一种适用于高动态强干扰环境的视觉辅助微机械捷联惯性导航系统/全球定位系统超紧组合导航系统[J]. 兵工学报, 2019, 40(11): 2241-2249.
[13]	刘生攀, 王文举, 饶兴桥. 单轴旋转捷联惯性导航系统误差分析与转位方案研究[J]. 兵工学报, 2018, 39(9): 1749-1755.
[14]	李伟，姜俊峰，任海波，张光辉. 行进间火炮指向稳定跟踪系统建模与控制[J]. 兵工学报, 2018, 39(8): 1479-1490.
[15]	王蕾，王玮，刘增军，宋天骁. 混合式惯性导航系统全数字平台技术研究[J]. 兵工学报, 2018, 39(7): 1316-1322.