Cubature Kalman Particle Filters

Abstract—To resolve the tracking problem of nonlinear/non-Gaussian systems effectively, this paper proposes a novel combination of the cubature kalman filter(CKF) with the particle filters(PF), which is called cubature kalman particle filters(CPF). In this algorithm, CKF is used to generate the importance density function for particle filter. It linearizes the nonlinear functions using statistical linear regression method through a set of Gaussian cubature points. It need not compute the Jacobian matrix and is easy to be implemented. Moreover, it makes efficient use of the latest observation information into system state transition density, thus greatly improving the filter performance. The simulation results are compared against the widely used unscented particle filter(UPF), and have demonstrated that CPF has higher estimation accuracy and less computational load. Keywords- particle filter; cubature kalman filter; importance density function I