Abstract:In multi-sensor multi-target bearings-only tracking systems,the measurements are usually unresolved,especially in cluttered environment.So tracking multi-target has to tackle the association between measurements and targets.In this paper,we proposed a new Rao-Blackwellized particle filter based on UKF for tracking multiple targets in cluttered environment.The RBPF-UKF allows the tracking algorithm to be partitioned into data association and single target tracking sub-problems,where the data association sub-problem can be solved by sequential importance resampling,and the single target tracking by UKF.Simulation result indicate that the RBPF-UKF can effectively resolve the data association sub-problem,and RBPF-UKF has better performance than RBPF-EKF in precision,stability when it is applied in bearing-only multi-target tracking.

Study on multi-sensor multi-target tracking using rao-blackwellized particle filter based on UKF