Fusion of possibly biased location estimates using Gaussian mixture models

A probabilistic framework for fusing location estimates, which may be biased and inconsistent, is presented. The proposed method, involving Gaussian mixture models (GMMs), utilizes prior information regarding the sensor bias, firstly, to reduce errors in the fused location estimate, and secondly, to produce a fused covariance matrix that better reflects the expected location error. Simulations are used to evaluate performance, relative to other techniques, such as the covariance union (CU) method. A passive geolocation application involving an airborne electronic support (ES) system is considered.

[1]  K. Spingarn.  Passive Position Location Estimation Using the Extended Kalman Filter , 1987, IEEE Transactions on Aerospace and Electronic Systems.

[2]  S. Challa,et al.  Joint sensor registration and track-to-track fusion for distributed trackers , 2004, IEEE Transactions on Aerospace and Electronic Systems.

[3]  H. Durrant-Whyte,et al.  Rich probabilistic representations for bearing only decentralised data fusion , 2005, 2005 7th International Conference on Information Fusion.

[4]  S. Rhodes.  Real time detection and compensation of registration errors in a network of radars , 2002 .

[5]  S.J. Davey,et al.  Asynchronous sensor registration via Expectation Maximisation , 2008, 2008 International Conference on Radar.

[6]  Barbara F. La Scala,et al.  Choosing a track association method , 2002, Inf. Fusion.

[7]  Jeffrey K. Uhlmann,et al.  A non-divergent estimation algorithm in the presence of unknown correlations , 1997, Proceedings of the 1997 American Control Conference (Cat. No.97CH36041).

[8]  Kutluyil Dogançay,et al.  On the bias of linear least squares algorithms for passive target localization , 2004, Signal Process..

[9]  Jeffrey K. Uhlmann,et al.  Covariance consistency methods for fault-tolerant distributed data fusion , 2003, Inf. Fusion.

[10]  P.N. Pathirana,et al.  Optimality Analysis of Sensor-Target Geometries in Passive Localization: Part 1 - Bearing-Only Localization , 2007, 2007 3rd International Conference on Intelligent Sensors, Sensor Networks and Information.

[11]  T. Kirubarajan,et al.  Multisensor multitarget bias estimation for general asynchronous sensors , 2005, IEEE Transactions on Aerospace and Electronic Systems.

[12]  Kuo-Chu Chang,et al.  Architectures and algorithms for track association and fusion , 2000 .

[13]  Carl G. Looney,et al.  Exploring fusion architecture for a common operational picture , 2001, Inf. Fusion.

[14]  S.J. Julier,et al.  A method for dealing with assignment ambiguity , 2004, Proceedings of the 2004 American Control Conference.

[15]  Hedvig Kjellström,et al.  An information fusion demonstrator for tactical intelligence processing in network-based defense , 2007, Inf. Fusion.

[16]  Jeffrey K. Uhlmann,et al.  Generalized Information Representation and Compression Using Covariance Union , 2006, 2006 9th International Conference on Information Fusion.

[17]  D. Nicholson.  An automatic method for eliminating spurious data from sensor networks , 2004 .