A Particle Filter Algorithm Based on Scaled UKF with Reduced Sigma Points
Guang Hui Zhao · Acta Automatica Sinica · 2015
In order to reduce the computation burden of conventional unscented particle filter(UPF), a method for particle filter based on minimal skew simplex unscented transformation(MSSUT) is proposed. This method uses a minimal skew simplex unscented Kalman filter to generate importance distribution of the particle filter. It can extend its overlaps and posterior probability density, and reduce the computation burden by reducing sigma points. However, the sigma point set coverage radius expands over dimension of state space, which results in the deterioration of the aggregation of sigma points. Auxiliary random variable formulation of the scaled transformation can overcome the defect of sigma point set distribution expansion. So a scaled minimal skew simplex unscented particle filter(SMSSUPF) is introduced. It is shown by simulation that compared with conventional unscented particle filter, the computation complexity of SMSSUPF can be reduced, the computation burden can be reduced, and compared with spherical simplex unscented particle filter(MSSUPF), SMSSUPF can reduce the system noise and the measurement noise variance estimation error.