基于一致性校验的贝叶斯估计器性能评估

系统仿真学报 ›› 2016, Vol. 28 ›› Issue (3): 569-576.

基于一致性校验的贝叶斯估计器性能评估

孙作雷¹, 李影¹, 张波², 张文¹

1.上海海事大学信息工程学院,上海 201306;
2.中国科学院上海高等研究所,上海 201210

收稿日期:2014-11-03 修回日期:2015-05-14 发布日期:2020-07-02
作者简介:孙作雷(1982-), 男, 山东枣庄, 博士, 副教授, 研究方向为移动机器人导航、机器学习等。
基金资助:
国家自然科学基金项目(61105097;51279098, 61401270);上海海事大学研究生创新基金(2015ycx059)

Performance Evaluation of Bayesian Estimator with Consistency Validation

Sun Zuolei¹, Li Ying¹, Zhang Bo², Zhang Wen¹

1. Information communication college, Shanghai Maritime University, Shanghai 201306, China;
2. Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, China

Received:2014-11-03 Revised:2015-05-14 Published:2020-07-02

摘要/Abstract

摘要： 针对贝叶斯估计器设计中缺乏严谨评估手段的问题,提出以估计一致性作为表征贝叶斯估计器性能的核心指标。通过分析观测更新中新息序列的无偏性和白特性,构造正则化新息平方序列,并对它实施卡方校验以判断相应贝叶斯估计器后验概率的估计一致性。仿真结果表明所提算法能同时在线性及非线性系统中有效辨识贝叶斯估计器设计中不恰当的参数配置,进而评估估计器性能。

关键词: 贝叶斯估计器, 一致性, 卡方检测, 正则化新息平方

Abstract: Considering the lack of sound evaluation method for Bayesian estimator design, an approach employing consistency as the core criterion was proposed. By studying the unbiasedness and whiteness of innovation sequence yielded during observation update routine, the normalized innovations square (NIS) was constructed, and then the chi-square test being able to detect the Posterior's consistency was carried out on NIS. Experiments demonstrate that the proposed algorithm can identify the inappropriate parameter configuration for both linear and nonlinear systems, therefore the performance of the estimators can be evaluated.

Key words: Bayesian estimator, consistency, chi-square test, normalized innovations squared

中图分类号:

TP13

孙作雷, 李影, 张波, 张文. 基于一致性校验的贝叶斯估计器性能评估[J]. 系统仿真学报, 2016, 28(3): 569-576.

Sun Zuolei, Li Ying, Zhang Bo, Zhang Wen. Performance Evaluation of Bayesian Estimator with Consistency Validation[J]. Journal of System Simulation, 2016, 28(3): 569-576.

参考文献

[1] Bergman N.Recursive Bayesian Estimation [D]. Sweden: Department of Electrical Engineering, Linköping University, Linköping Studies in Science and Technology, 1999: 579-586.
[2] Zheng Y, Niu R, Varshney P K.Sequential Bayesian estimation with censored data for multi-sensor systems[J]. Signal Processing, IEEE Transactions on (S0165-1684), 2014, 62(10): 26-41.
[3] Ali-Löytty S, Sirola N, Piché R.Consistency of three Kalman filter extensions in hybrid navigation[C]// Proceedings of The European Navigation Conference on GNSS, Munich, Germany. Finland: Tampereen teknilline yliopisto (Tampere University of Technology), 2005: 1049-1056.
[4] Lei M, van Wyk B J, Qi Y. Online estimation of the approximate posterior cramer-rao lower bound for discrete-time nonlinear filtering[J]. Aerospace and Electronic Systems, IEEE Transactions on (S0885-8985), 2011, 47(1): 37-57.
[5] Kaess M, Johannsson H, Roberts R, et al.iSAM2: Incremental smoothing and mapping with fluid relinearization and incremental variable reordering[C]// Proceedings of the Robotics and Automation (ICRA), 2011 IEEE International Conference on. USA: IEEE, 2011: 32(8): 1-8.
[6] Huang G P, Mourikis A I, Roumeliotis S I.Observability-based rules for designing consistent EKF SLAM estimators[J]. The International Journal of Robotics Research (S0278-3649), 2010, 29(5): 20-28.
[7] Sussman D L, Tang M, Priebe C E.Consistent latent position estimation and vertex classification for random dot product graphs[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on (S0162-8828), 2014, 36(1): 48-57.
[8] Dudek G, Jenkin M.Computational principles of mobile robotics [M]. UK: Cambridge University Press, 2010.
[9] Schlosser M S, Kroschel K.Limits in tracking with extended Kalman filters[J]. Aerospace and Electronic Systems, IEEE Transactions on (S0018-9251), 2004, 40(4): 1-9.
[10] Lu F, Milios E.Robot pose estimation in unknown environments by matching 2d range scans[J]. Journal of Intelligent and Robotic Systems (S0921-0296), 1997, 18(3): 49-75.