A State Estimation Method Based on Interactive Multiple Model Particle Filter

XIAO Yanghui,SHI Zelin,ZHAO Yongting
DOI: https://doi.org/10.3969/j.issn.1673-9728.2011.06.052
2011-01-01
Abstract:In order to improve the state estimation performance which belongs to target tracking,the interactive multiple model particle filter(IMMPF) adopts IMM to describe target movement,and estimate the target state by particle filters.IMMPF can avoid estimation error caused by single motion model.It also overcomes the limitation of Kalman tracking filter.When the priori knowledge is inadequate,IMMPF can estimate target position,velocity and acceleration with appropriate error under different form of exercise.So IMMPF improves adaptability and robustness.
What problem does this paper attempt to address?