Robust IMM Filtering Approach with Adaptive Estimation of Measurement Loss Probability for Surface Target Tracking

A suitable jump Markov system (JMS) filtering approach provides an efficient technique for tracking surface targets. In complex target situations, due to the joint influences of lost measurements with unknown probability and heavy-tailed measurement noise (HTMN), estimation accuracy conventional interacting multiple model (IMM) methods may be seriously degraded. Aiming address issues in JMSs HTMNs random losses, this paper presents IMM adaptive loss probability. study, we assumed that noises obey student’s t-distributions then proposed Bernoulli variables (BRVs) characterize loss. Notably, by converting two likelihood functions from weighted sum form exponential multiplication, established hierarchical Gaussian state space models directly utilize variational inference method. The vectors, distribution parameters, BRVs, probabilities were estimated simultaneously according Bayesian framework. results maneuvering simulations verified presented demonstrated superior compared existing filters.