KLD-Sampling-Based Adaptive Unscented Particle Filter for Nonlinear and Non-Gaussian State Estimation

摘要：

Particle Filter (PF) is well known as a state estimation method for nonlinear and non-Gaussian system.However,PF has the inherent drawbacks such as samples less of diversity and the calculation complexity of PF depends on the number of samples used for state estimation process.In this paper,the Adaptive Unscented Particle Filter (AUPF) is proposed in order to overcome these drawbacks.In the new algorithm,the KLD-sampling and Unscented Kalman Filter (UKF) are simultaneously used to improve the performance of particle filter.The new algorithm overcomes the drawback of cost of calculation by adapting the size of sample sets through KLD-sampling during the estimation process.The new algorithm overcomes drawback of less of diversity using UKF to calculate the important sample probability during the estimation process.The computer simulations are performed to compare the AUPF algorithm,the traditional UPF and PF in performance.The simulation results demonstrated that the AUPF is very efficient and smaller time consumption compared to traditional UPF and PF.Therefore,the AUPF is more suitable to the nonlinear and non-Gaussian state estimation.

关键词： Adaptive Unscented Particle Filter KLD- Sampling Nonlinear and Non-Gaussian State Estimation

作者: Fujun Pei Pingyuan Cui Yangzhou Chen

作者单位: School of Electronic Information＆Control Engineering Beijing University of Technology,Beijing,China

会议类型: 国际会议

会议名称: International Conference on Modelling,Identification and Control(模拟、鉴定、控制国际会议)

会议地点: 上海

会议语种:英文

在线出版日期: 2008-06-29（万方平台首次上网日期，不代表论文的发表时间）