Copula-based particle filtering for target tracking in non-linear/non-Gaussian scenarios with correlated sensors

One of the most important challenges in target tracking is the modeling of correlated and non-Gaussian disturbance. In this paper, the problem of target tracking by the means of particle filtering in environments with highly correlated sensors, is speculated. The goal is to prepare an accurate model of dependency structure in multivariate observation likelihood function, with non-Gaussian marginals to achieve a new algorithm in tracking problems. The main novelty of our method, termed as Copula-based Sequential Importance Resampling particle filter (CSIR-PF), is the application of the copula theory which is a powerful tool in correlation modeling in statistical theory. The precise model obtained by copulas, makes a great improvement in particles' weighting. The performance of our proposed method is depicted through simulations of problems with highly correlated sensors. Results clearly indicate the acceptable performance of CSIR-PF.