Cubature Particle Filter and Its Application
Junmin Zhang · Xi'an Jiaotong Daxue xuebao · 2011
A new particle filter named cubature particle filter(CPF) is proposed to improve the low state estimation accuracy of existing particle filters for nonlinear/non-Gaussian discrete time systems.The CPF directly uses the cubature rule based numerical integration method to calculate the mean and covariance,to generate the proposal density function for the particle filter,and to obtain the required particles with weights.Then the minimum square error state estimation is obtained based on these particles and weights.The particles generated using the CPF algorithm involves the use of the latest measurements so that the approximation to the system posterior density is improved.Simulation results show that the estimation error of the CPF algorithm is about one-fifth of that of the generic particle filter and one-third of that of the extended particle filter,respectively,and half of that of unscented particle filter(UPF),while the run time of CPF is only one third of that of UPF.