A FISST Approach for Single Target Multi-measurement Problem

摘要

Inspired by the current popular multi-target Probability Hypothesis Density (PHD) filter, this paper uses Random Finite Sets (RFS) formalism to solve the problem of tracking a target that can be observed by multi-sensor. The solution to the single target multi-measurement problem is built on Finite Set Statistics (FISST) Bayesian framework. In addition, a SMC implementation of the FISST filtering is proposed and demonstrated on a number of particles. Compared with the traditional methods, such as the Extend Kalman Filter (EKF) and Particle Filter (PF), FISST filtering prove to be best in tracking.