Research on Particle Filter Based on an Improved Hybrid Proposal Distribution with Adaptive Parameter Optimization

摘要

Although it has attracted widespread attentions in the nonlinear filtering field, particle filter algorithm exists the sample degradation problem. In order to improve the algorithm performance, an improved hybrid proposal distribution with adaptive parameter optimization for particle filter is studied. Firstly, based on the performance analysis of different proposal distribution, a hybrid proposal distribution with fixed annealing parameter (called improved hybrid proposal distribution) is utilized to consider current information of the latest observed measurement. Then, aimed at the deficiency about annealing parameter using fixed value, improved hybrid proposal distribution with adaptive optimization of annealing parameter is proposed by comparison with the relationship among true state, observational data and forecast data based on proposal distribution. With the simulation program, the performance of the proposed strategy is evaluated and its validity is verified.