Robust centralized multi-sensor fusion using cubature information filter

Noise statistics is crucial to the estimation fusion problem for the sensor network. In this paper, a centralized multi-sensor fusion for the network with non-Gaussian measurement noises is considered. The heavy-tailed Student-t distribution is chosen to model the non-Gaussian measurement noise for each sensor. By using a hierarchical model for the Student-t distribution, Wishart distribution and Gamma distribution are introduced to model the uncertainties of the noise parameters. The recursive update equations for the combined information state, combined information matrix, and noise parameters are derived by using variational Bayesian approach together with spherical cubature integration handling nonlinearities. Specifically, the information state and information matrix are obtained based on the fused local estimates of non-Gaussian noise parameters from all the sensors. A multi-sensor target tracking example illustrates the robustness of the proposed multi-sensor fusion method to the outliers.

[1]  Yu Hen Hu,et al.  Generalised Kalman filter tracking with multiplicative measurement noise in a wireless sensor network , 2014, IET Signal Process..

[2]  Simo Särkkä,et al.  Non-linear noise adaptive Kalman filtering via variational Bayes , 2013, 2013 IEEE International Workshop on Machine Learning for Signal Processing (MLSP).

[3]  Christopher M. Bishop,et al.  Robust Bayesian Mixture Modelling , 2005, ESANN.

[4]  Henry Leung,et al.  A variational Bayesian approach to robust sensor fusion based on Student-t distribution , 2013, Inf. Sci..

[5]  Wenqiang Liu,et al.  Robust centralized fusion Kalman predictors with multiplicative noises, uncertain noise variances, and missing measurements , 2016, 2016 Chinese Control and Decision Conference (CCDC).

[6]  Zhi Tian,et al.  Performance evaluation of track fusion with information matrix filter , 2002 .

[7]  Junping Du,et al.  Distributed filtering for discrete-time linear systems with fading measurements and time-correlated noise , 2017, Digit. Signal Process..

[8]  V. Šmídl,et al.  The Variational Bayes Method in Signal Processing , 2005 .

[9]  Simo Särkkä,et al.  Recursive outlier-robust filtering and smoothing for nonlinear systems using the multivariate student-t distribution , 2012, 2012 IEEE International Workshop on Machine Learning for Signal Processing.

[10]  Deok-Jin Lee,et al.  Nonlinear Estimation and Multiple Sensor Fusion Using Unscented Information Filtering , 2008, IEEE Signal Processing Letters.

[11]  Quanbo Ge,et al.  Information fusion algorithm based on adaptive cubature strong tracking filter , 2017, 2017 36th Chinese Control Conference (CCC).

[12]  Matthew J. Beal.  Variational algorithms for approximate Bayesian inference , 2003 .

[13]  Yingmin Jia,et al.  Distributed consensus filtering for discrete-time nonlinear systems with non-Gaussian noise , 2012, Signal Process..

[14]  Ian Postlethwaite,et al.  Cubature information filter and its applications , 2011, Proceedings of the 2011 American Control Conference.

[15]  Masa-aki Sato,et al.  Online Model Selection Based on the Variational Bayes , 2001, Neural Computation.

[16]  Chen Shen,et al.  A Robust Particle Filtering Algorithm With Non-Gaussian Measurement Noise Using Student-t Distribution , 2014, IEEE Signal Processing Letters.

[17]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[18]  文成林,et al.  Cubature Information Filters with Correlated Noises and Their Applications in Decentralized Fusion , 2014 .

[19]  Meng Cai,et al.  Multi-sensor adaptive data fusion with colored measurement noise , 2017, 2017 29th Chinese Control And Decision Conference (CCDC).