Comparison of Particle Filter and Extended Kalman Particle Filter
Jacek Michalski, Piotr Kozierski, Joanna Ziętkiewicz · CeON Repository (Centre for Evaluation in Education and Science) · 2017
In this paper, three state estimation algorithms, namely: Extended Kalman Filter, Particle Filter (Bootstrap Filter) and Extended Kalman Particle Filter, have been presented. Particle Filter and Extended Kalman Particle Filter algorithms have been compared with a different number of particles and the results have been presented together with Extended Kalman Filter. Estimation quality has been checked for three nonlinear objects (one- and multidimensional systems) and evaluated through the aRMSE quality index value. Based on the obtained results it was concluded that Extended Kalman Particle Filter provide better estimation quality for low number of particles in comparison to simple particle filter. However it is not met for highly nonlinear system.