Performance Analysis of Kalman-Based Filters and Particle Filters for Non-Linear/Non-Gaussian Bayesian Tracking

Authors:	Shu Wenjie, National University of Defense Technology, China Zheng Zhiqiang, National University of Defense Technology, China
Topic:	1.1 Modelling, Identification & Signal Processing
Session:	State Estimation, Tracking and Sensor Fusion
Keywords:	Kalman-based filters, Particle filters, Bayesian tracking

Abstract

In this paper, we present an overview performance analysis ofKalman-based filters and particle filters forNon-Linear/Non-Gaussian Bayesian tracking. The simulation resultsshow that the particle filters has superior performance than theKalman-based filters. Although the particle filters is timeconsuming, but in many situations such as the low data rate, lowsignal-to-noise ratio situations, the superior performance is veryattractive.