Average Weight Optimization RBPF Method for Target Tracking in Multi-Sensor Observations

The reasonable design of particle filter framework in multi-sensor observation system is the key to expand the application domain of sampling nonlinear filters. Aiming at the effective realization of particle filter for multi-sensor target tracking problem, a novel average weight optimization Rao-Blackwellised particle filtering algorithm is proposed. Combining with the kinetic equation of target state evolution, RBPF is used as the basic estimator of algorithm realization. For the rational utilization from multi-sensor observations and the reduction of the adverse influence from random observations noise in measuring process of particles weight, the average weight optimization strategy is used to improve the reliability and stability of particle weight variance. In addition, we give the concrete flow of RBPF in average weight optimization strategy. Finally, the theoretical analysis and experimental results show the feasibility and efficiency of the proposed algorithm.