Distributed hybrid consensus–based square-root cubature quadrature information filter and its application to maneuvering target tracking

Abstract

To handle nonlinear filtering problems with networked sensors in a distributed manner, a novel distributed hybrid consensus–based square-root cubature quadrature information filter is proposed. The proposed hybrid consensus–based square-root cubature quadrature information filter exploits fifth-order spherical simplex-radial quadrature rule to tackle system nonlinearities and incorporates a novel measurement update strategy into the hybrid consensus filtering framework, which takes the predicted measurement error into account and hence produces more accurate estimates. In addition, the proposed hybrid consensus–based square-root cubature quadrature information filter inherits the complementary positive features of both consensus on information and consensus on measurements methods and avoids sensitive matrix operations such as square-root decompositions and inversion of covariances, which is beneficial for numerical stability. Stability analysis with respect to consensus, convergence, and consistency for the proposed hybrid consensus–based square-root cubature quadrature information filter is also developed. The effectiveness of the proposed hybrid consensus–based square-root cubature quadrature information filter is validated through a maneuvering target tracking scenario. The simulation results show that the proposed hybrid consensus–based square-root cubature quadrature information filter outperforms the existing algorithms at the expense of a slight increase in computational cost.

Distributed square root fuzzy cubature information filter for object tracking on the possibilistic framework

“…The corresponding prior error and prediction estimate error are expressed as $${\widehat{\eta }}_{i,t\vert t}\triangleq {\tilde{\widehat{x}}}_{i,t\vert t}-{x}_{t}$$ and $${\widehat{\eta }}_{i,t+1\vert t}\triangleq {\tilde{\widehat{x}}}_{i,t+1\vert t}-{x}_{t}$$ respectively, where $${\tilde{\widehat{x}}}_{i,t\vert t}=\mathbb{C}\left[{\widehat{x}}_{i,t\vert t}\right]$$, $${\tilde{\widehat{x}}}_{i,t+1\vert t}=\mathbb{C}\left[{\widehat{x}}_{i,t+1\vert t}\right]$$, x t is the unknown real state. The pseudo transition matrix can be computed using the statistical linear error propagation methodology [5] as $$${F}_{i,t}\approx {P}_{i,\left({x}_{t\vert t},{x}_{t+1\vert t}\right)}^{\mathrm{T}}{Y}_{i,t\vert t},$$$ where $${P}_{i,\left({x}_{t\vert t},{x}_{t+1\vert t}\right)}$$…”

“…Remark We have not made simulation comparisons between DSRFCIF and the distributed square root cubature information filtering [5] on the probability framework. It is because the possibility method does not always perform better than the probability method.…”

“…In particular, the objective of this paper is to formulate a stable fuzzy distributed filtering for a non-linear system with fuzzy noise to ensure that, each sensor i can increase its local posterior estimation E xi;t|t 6) and non-linear motion Equation (5).…”

“…x t is the unknown real state. The pseudo transition matrix can be computed using the statistical linear error propagation methodology [5] as…”

“…However, numerically sensitive matrix operations in DCF often cause troubles. To avoid this problem, distributed square root cubature filter (DSRCF) was proposed [5]. Furthermore, scholars have also conducted a lot of research on specific scenarios.…”