Improved particle filter algorithms based on partial systematic resampling

As a hot research topic, particle filter (PF), has been successfully applied into many fields. Combined with the analysis of partial stratified resampling (PSR) algorithm, two kinds of improved PF algorithm are presented. One improved PF algorithm with weights optimization is to use the optimal idea to improve the weights after implementing PSR resampling so as to enhance the performance of PF. The other PF algorithm based on adaptive mutation resampling is also to use the weights optimal idea for dominant or negligible particles in order to improve the resampling performance before implementing PSR resampling; and used the mutation operation for all particles so as to assure the diversity of particle sets. At the same time, the adaptive resampling mechanism is introduced to improve the performance of PF. At last, with the simulation program using matlab 7.0 to track a single target motion from a fixed visual observation points, the performance of the proposed algorithm is evaluated and its validity is verified.