基于ESMS过滤器的信息融合理论研究及SLAM应用

本文解决了多源信息融合时信息源选择的难题，提出了一种广义的证据支持贴近度过滤器来选择最一致的证据源，并耦合基于DSmT和PCR5的融合机，应用于Pioneer Ⅱ移动机器人的SLAM;通过对运行在虚拟环境中的一个虚拟机器人（自身携带16个Sonar传感器），感知周围环境信息，对有或没有ESMS过滤器两种情况下的环境地图重构效果进行比较，充分验证了ESMS过滤器作为信息融合源选择先决条件的优点。

Research and SLAM Application of Information Fusion Theory Based on ESMS Filter

In this paper, we address the problem of selection of sources of information as a prerequisite for the fusion, and propose a very general evidence supporting measure of similarity (ESMS)for selecting the most coherent subset of sources to combine among all sources available at each instant. ESMS filter coupling with the fusion machine based on DSmT and PCR5 is applied to the SLAM of the Pioneer II mobile robot, where a virtual automatic mobile robot detecting the environment from 16 sonar sensors evolves in the virtual environment with obstacles. By comparing the result of map reconstruction of the world with or without ESMS filter, it testifies the superiority of the selection of the sources as prerequisite for improvement of information.