Research on particle filter algorithms based on cost reference

YAO Yuqing，HE Tongneng，CHEN Peijun
（College of Information Engineering， Zhejiang University of Technology， Hangzhou 310023， China）

Abstract：In connection with the problem on nonlinear model filter， costreference particle filtering and standard articles filtering were researched respectively. The differences and similarities between algorithm structure and algorithm were analyzed. The effects of the two filter in the same singlevariable nonstatic growth model were analyzed and compared by  Matlab. The research results indicate that the filter based on the mathematical statistic is better than the filter based on the probability statistics for solving the nonlinear problem.
Key words：nonlinear problem; particle filter; reference costs function