A new robust distributed IMM filter for tracking unmanned aerial vehicle based on multivariate Laplace distribution

In the process of tracking manoeuvring target unmanned aerial vehicles (UAV), the appearance of heavy-tailed measurement noise (HTMN) evoked by outliers gives rise to a decrease in the estimation accuracy of traditional filtering algorithms and even divergence. To deal with this problem, a new robust distributed interacting multiple mode (IMM) using weighted average consensus (WAC) based on multivariate Laplace distribution (MLD) is designed. Firstly, the measurement noise is modelled as MLD, the inverse Wishart (IW) distribution is chosen as the conjugate prior distribution of scale matrices. Secondly, the system state vector and noise scale matrices are jointly inferred using variational Bayesian (VB) technique. What's more, information pairs and model probabilities are modified employing WAC to improve estimate performance of the devised algorithm and enhance the reliability of the sensor network. Finally, the effectiveness of the devised algorithm is illustrated through a simulation experiment.