文档介绍：关于变压器的论文关于传感器的论文
粗糙集多传感器融合理论在大棚喷雾机器人中的应用
摘要:针对多传感器融合过程中传感器数据过载的问题,提出了一种基于粗糙集约简理论的方法,应用于大棚喷雾机器人植物识别系统之中。首先根据植物的特性,建立植物种类识别信息表即初始决策表,然后根据粗糙集约简规则推导出最小决策规则,最后得到杂草识别区分的最快融合算法。
关键词:粗糙集;多传感器融合;机器人
Rough Set and Multi-sensor Fusion Theory and Its Application in Greenhouse
Spray Robot
LIANG Wei,XIAO Ai-ping,QIAN Hua
(School of Technology, Beijing Forestry University, Beijing 100083, China)
Abstract: When syncretizing multi-sensor information, the sensor datas were often overloaded. The method based on rough set reducing decision-making and deleting redundant attributes was proposed for plant identification system of greenhouse spray robot. Firstly,according to the characteristics of plants, we established plant species identification information table(the initial decision-making table).Secondly, the minimum decision-making rules were deducted based on rough set reduction rules. Finally, we got the fastest fusion algorithm of weed identification and distinction.
Key words: rough set; multi-sensor fusion; robot
1粗糙集与多传感器融合理论简介
多传感器信息融合的实质是对多源不确定性信息的分析与综合[1],其处理过程非常复杂,在其融合过程中,我们经常会遇到传感器数据的过载问题,例如在用神经网络进行传感器信息融合时存在如何选择样本集的问题。这些问题制约了传感器信息融合的发展,粗糙集理论的出现,对解决这些问题提供了具体可行的思路。
[2,3]。目前,粗糙集理论已被成功地运用于机器学****过程控制、决策分析、专家系统与智能控制、数据挖掘等研究领域[4]。
目前,智能机器人在进行物体辨别的时候,首先需要根据物体的基本特征,根据经验知识等,建立物体的特征数据库,然后再将传感器测得的值与特征数据库进行匹配,最后得出匹配结果[