Joint estimation of state and system biases in non-linear system

In multi-platform surveillance system, a prerequisite for successful fusion is the transformation of data from different platforms to a common coordinate system. However, some stochastic system biases arise during this transformation, and they seriously downgrade the global surveillance performance. Considering that the target state and the system biases are coupled and interactive, the authors present a new recursive joint estimation (RJE) algorithm for registering stochastic system biases and estimating target state. First, the relationship between system biases estimation and target state estimation is derived. Second, the RJE framework is introduced on the basis of the proposed relationship. Representing the different behavioural aspects of the motion of a maneuvering target is difficult to achieve with a single model in a multi-platform target tracking system. By accounting for the non-linear and/or non-Gaussian property of the dynamic system, they modify the interacting multiple model–particle filter framework to estimate parameters. This approach considers not only the influence of the system biases, but also the covariance of state on the basis of multiple-particle statistics. Simulation results reveal the superior performance of the proposed approach with respect to the traditional algorithm under the same conditions.

[1]  Ali T. Alouani,et al.  Sensor registration in multisensor systems , 1992, Defense, Security, and Sensing.

[2]  Y. Bar-Shalom,et al.  Exact multisensor dynamic bias estimation with local tracks , 2004, IEEE Transactions on Aerospace and Electronic Systems.

[3]  Y. Bar-Shalom,et al.  Multisensor target tracking performance with bias compensation , 2006, IEEE Transactions on Aerospace and Electronic Systems.

[4]  Henry Leung,et al.  An exact maximum likelihood registration algorithm for data fusion , 1997, IEEE Trans. Signal Process..

[5]  I. T. Li,et al.  Multi-target multi-platform sensor registration in geodetic coordinates , 2002, Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997).

[6]  James P. Reilly,et al.  An EM Algorithm for Nonlinear State Estimation With Model Uncertainties , 2008, IEEE Transactions on Signal Processing.

[7]  B. Ristic,et al.  Maximum likelihood registration for multiple dissimilar sensors , 2003 .

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

[9]  P. H. Foo,et al.  Combining the interacting multiple model method with particle filters for manoeuvring target tracking , 2011 .

[10]  Y. Boers,et al.  Interacting multiple model particle filter , 2003 .

[11]  Henry Leung,et al.  An EM-IMM Method for Simultaneous Registration and Fusion of Multiple Radars and ESM Sensors , 2010 .

[12]  H. Leung,et al.  Space-time registration of radar and ESM using unscented Kalman filter , 2004, IEEE Transactions on Aerospace and Electronic Systems.

[13]  B. Friedland.  Treatment of bias in recursive filtering , 1969 .

[14]  William Dale Blair,et al.  Interacting multiple bias model algorithm with application to tracking maneuvering targets , 1992, [1992] Proceedings of the 31st IEEE Conference on Decision and Control.

[15]  P.M. Djuric,et al.  Bearings-Only Tracking with Biased Measurements , 2007, 2007 2nd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing.

[16]  Yifeng Zhou,et al.  Sensor alignment with Earth-centered Earth-fixed (ECEF) coordinate system , 1999 .

[17]  Henry Leung,et al.  Joint Data Association, Registration, and Fusion using EM-KF , 2010, IEEE Transactions on Aerospace and Electronic Systems.

[18]  Amir Averbuch,et al.  Interacting Multiple Model Methods in Target Tracking: A Survey , 1988 .

[19]  M. Ignagni.  An alternate derivation and extension of Friendland's two-stage Kalman estimator , 1981 .

[20]  S. Dhar.  Application of a Recursive Method for Registration Error Correction in Tracking with Multiple Sensors , 1993, 1993 American Control Conference.