基于逻辑回归的总系统误差分类和预测方法

doi:10.16265/j.cnki.issn 1003-3033.2020.12.015

中国安全科学学报 ›› 2020, Vol. 30 ›› Issue (12): 106-112.doi: 10.16265/j.cnki.issn 1003-3033.2020.12.015

基于逻辑回归的总系统误差分类和预测方法

焦卫东教授, 王维

中国民航大学天津市智能信号与图像处理重点实验室,天津 300300

收稿日期:2020-09-20 修回日期:2020-11-15 出版日期:2020-12-28 发布日期:2021-07-15
作者简介:焦卫东 (1973—),男,陕西咸阳人,博士,教授,主要从事飞行程序设计、图像/视频处理与编码、三维虚拟仿真验证等方面的研究。E-mail: nxjiaowd@sina.com。
基金资助:
国家自然科学基金资助(U1533115)。

Total system error classification and prediction based on logistic regression

JIAO Weidong, WANG Wei

Tianjin Key Lab for Advanced Signal Processing, Civil Aviation University of China, Tianjin 300300, China

Received:2020-09-20 Revised:2020-11-15 Online:2020-12-28 Published:2021-07-15

摘要/Abstract

摘要： 为保证飞行总系统误差(TSE)超出安全容限时机载告警系统能及时发出正确告警,提出以影响TSE的因素为逻辑回归法(LRM)的样本特征,利用LRM进行TSE分类和预测的方法,快速将TSE预判为超限和不超限;选取对TSE影响比较稳定的影响因素,如导航系统误差(NSE)、位置精度因子(PDOP)值及卫星可见数目等为样本特征,开展试验;分析不同种类和数量的样本特征及组合下LRM对TSE分类和预测准确性的影响,并与传统基于坐标计算的TSE预测方法的准确率和计算负荷作比较。结果表明:以NSE、PDOP值和卫星可见数为组合,LRM对TSE的预测准确率最高,计算负荷相对较小,且均优于基于坐标计算的TSE预测方法。

关键词: 逻辑回归法(LRM), 总系统误差(TSE), 导航系统误差(NSE), 位置精度因子(PDOP), 卫星可见数

Abstract: In order to ensure that airborne warning system can send out correct alarm in time when TSE of flights is beyond safety limit, a LRM-based method to classify and predict TSE as in limit or over limit was proposed with influencing factors of TSE as sample characteristics of LRM. Experiments were conducted with stable factors, such as NSE, PDOP value and visible number of satellites selected as sample characteristics. Then, influence of LRM on classification and prediction accuracy of TSE was analyzed under conditions of different types and quantities of sample characteristics and combinations, and prediction accuracy and computational load were compared with that of traditional TSE estimation methods based on coordinate calculation. The results show that in the event of a combination of NSE, PDOP value and satellite visible number, LRM has highest prediction accuracy for TSE, and its calculation load is relatively small, which is better than that of TSE prediction methods based on coordinate calculation.

Key words: logistic regression method (LRM), total system error (TSE), navigation system error (NSE), position dilution of precision (PDOP), visible satellites number

中图分类号:

X949

焦卫东, 王维. 基于逻辑回归的总系统误差分类和预测方法[J]. 中国安全科学学报, 2020, 30(12): 106-112.

JIAO Weidong, WANG Wei. Total system error classification and prediction based on logistic regression[J]. China Safety Science Journal, 2020, 30(12): 106-112.

参考文献

[1] 郝亮. PBN运行飞行程序设计研究[D]. 成都:电子科技大学, 2011.
HAO Liang. PBN operational flight program design study[D]. Chengdu: University of Electronic Science and Technology of China,2011.
[2] International Civil Aviation Organization.Performance-based navigation (PBN) manual[Z]. 2008.
[3] 中国民用航空局.中国民航基于性能的导航实施技术路线图 [Z].2009-10.
[4] ZHAO Xin, WANG Shicheng, ZHANG Jinsheng, et al. Real-time fault detection method based on belief rule base for aircraft navigation system[J]. Chinese Journal of Aeronautics, 2013, 26(3): 717-729.
[5] WOLFGANG S, BAI Jie, FENG Shaojun, et al. Integrity monitoring algorithms for airport surface movement[J]. Gps Solutions, 2012, 16(1): 65-75.
[6] 韩松臣, 裴成功, 隋东, 等. 平行区域导航航路安全性分析[J]. 航空学报, 2006, 27(6):1 023-1 027.
HAN Songchen, PEI Chenggong, SUI Dong, et al. Security analysis of area navigation parallel airway[J].Acta Areonautica ET Astronautica Scinica,2006, 27(6):1 023-1 027.
[7] 李楠, 强懿耕, 孙瑜, 等. 终端区航空器异常轨迹识别研究[J]. 中国安全科学学报, 2018, 28(11): 25-31.
LI Nan, QIANG Yigeng, SUN Yu, et al. Research on identification of aircraft abnormal trajectory in terminal area[J]. China Safety Science Journal, 2018, 28(11): 25-31.
[8] KIM E. A coverage analysis methodology for APNT pseudolite ground network[C]. Proceedings of the International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS), 2011:495-507.
[9] STAPLETON D P. A comparison of navigation by basic GPS lateral guidance to navigation by GPS/WAAS for various aircraft final approach categories[C]. Proceedings of the International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS), 2005: 629-642.
[10] FU Li, ZHANG Jun, LI Rui. Real-time total system error estimation: modeling and application in required navigation performance[J]. Chinese Journal of Aeronautics, 2014, 27(6): 1 544-1 553.
[11] 郭婧, 薛广月, 王嫣然. RNP运行中的实际导航性能评估技术研究[J]. 民航学报, 2018,2(3): 29-33.
GUO Jing, XUE Guangyue, WANG Yanran. Research on actual navigation performance evaluation algorithm for RNP operation[J]. Journal of Civil Aviation, 2018,2(3): 29-33.
[12] CASSELL R, SMITH A. Development of required navigation performance (RNP) requirements for airport surface movement guidance and control[C]. Digital Avionics Systems Conference, 1995:57-64.
[13] RICHARDSON D W. Derminnation of the impact of digital data broadcast on flight technical error [R]. NTIS, 1980.
[14] ZHAO Hongsheng, XU Xiaohao, ZHANG Jun, et al. Lateral flight technical error estimation model for performance based navigation[J]. Chinese Journal of Aeronautics, 2011, 24(3): 329-336.
[15] 赵鸿盛, 徐肖豪, 李锐, 等. 性能基导航中飞行技术误差估计研究综述[J]. 交通运输工程学报, 2014,14(5): 101-110.
ZHAO Hongsheng, XU Xiaohao, LI Rui, et al. Overview of research on flight technical error estimation in performance based navigation[J]. Journal of Traffic and Transportation Engineering, 2014,14(5): 101-110.
[16] HALL T, SOARES M. Analysis of localizer and glide slope flight technical error[C]. 2008 IEEE/AIAA 27^th Digital Avionics Systems Conference, 2008: 1-9.
[17] BRADLEY R, MACLAREN W M. Ordinal logistic regression analysis of flight task ratings[J]. The Aeronautical Journal, 2006, 110(1109): 447-456.
[18] PAVLYSHENKO B. Machine learning, linear and Bayesian models for logistic regression in railure detection problems[C]. IEEE International Conference on Big Data, 2017: 2 046-2 050.
[19] 毛林, 陆全华, 程涛. 基于高维数据的集成逻辑回归分类算法的研究与应用[J]. 科技通报, 2013, 29(12): 64-66.
MAO Lin, LU Quanhua, CHENG Tao. The research and application of ensemble logistic regression classification algorithm based on high dimensional data[J]. Bulletin of Science and Technology, 2013, 29(12): 64-66.
[20] 李琼阳. 基于再缩放策略的逻辑回归算法及其应用[J]. 统计与决策, 2019(10): 72-74.
LI Qiongyang. Rescaling strategy-based logistic regression algorithm and its application[J]. Statistics & Decision, 2019(10): 72-74.

基于逻辑回归的总系统误差分类和预测方法

Total system error classification and prediction based on logistic regression

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	何昕, 马义龙. 基于计算流体力学的配对进近尾流安全间隔研究[J]. 中国安全科学学报, 2022, 32(5): 77-83.
[2]	岳仁田, 张知波. 脆弱性多因素耦合作用下空管亚安全态识别[J]. 中国安全科学学报, 2022, 32(4): 8-14.
[3]	孙瑞山, 陈毅, 孙立斌. 基于CFF的管制员疲劳评估方法[J]. 中国安全科学学报, 2022, 32(4): 192-197.
[4]	吕旭飞, 姚尚宏, 权家乐. 基于STPA法的空中加油软管甩鞭安全性分析^*[J]. 中国安全科学学报, 2022, 32(2): 152-157.
[5]	王兴隆, 陈仔燕, 刘岩. 基于组合赋权法的空管监视信息质量评估^*[J]. 中国安全科学学报, 2022, 32(1): 41-50.
[6]	韩鹏, 王军, 王启, 赵嶷飞. 基于飞行剖面的航空器试飞活动空管风险研究^*[J]. 中国安全科学学报, 2022, 32(1): 149-156.
[7]	焦卫东, 姚军强, 王瑞冬. 基于凸包围盒的飞行冲突检测算法^*[J]. 中国安全科学学报, 2021, 31(12): 32-38.
[8]	白鹏, 陈霖峰, 王玺, 龚宇, 蒋淞榕, 周勍琪. 基于点融合的进场程序优化与航空器排序研究^*[J]. 中国安全科学学报, 2021, 31(12): 45-52.
[9]	曹惠玲, 胡彦杰. 适航指令设计制造缺陷案例关联分析^*[J]. 中国安全科学学报, 2021, 31(11): 54-63.
[10]	魏志强, 李晓晨. 飞机高空尾涡安全评估模型^*[J]. 中国安全科学学报, 2021, 31(11): 78-84.
[11]	杨晓强, 林湘漪, 蒋澜. 使用困难报告数据驱动的民用航空器安全状态评价^*[J]. 中国安全科学学报, 2021, 31(11): 85-92.
[12]	康瑞, 杨凯. 航空器逆向滑行冲突风险评估模型^*[J]. 中国安全科学学报, 2021, 31(10): 39-45.
[13]	吴仁彪, 何宇翔, 贾云飞. 基于改进层次分析法的重点人员风险评价方法^*[J]. 中国安全科学学报, 2021, 31(10): 112-118.
[14]	冯文刚. 基于深度长短记忆模型的民航安保事件分析^*[J]. 中国安全科学学报, 2021, 31(9): 1-7.
[15]	肖国松, 刘嘉琛, 董磊, 孙紫荆, 赵长啸. 面向IMA通用系统管理的STPA安全性分析^*[J]. 中国安全科学学报, 2021, 31(9): 8-14.