基于决策树的羽流追踪机器人自主决策方法研究  

Research on Autonomous Decision-Making of Plume Tracking Robots Using Decision Tree

作　　者：赵攀 袁杰[1] 王宏伟 米汤[1] ZHAO Pan;YUAN Jie;WANG Hongwei;MI Tang(School of Electrical Engineering,Xinjiang University,Urumqi 830047,China;School of Control Science and Engineering,Dalian University of Technology,Dalian,Liaoning 116024,China)

机构地区：[1]新疆大学电气工程学院,乌鲁木齐830047 [2]大连理工大学控制科学与工程学院,辽宁大连116024

出　　处：《计算机工程与应用》2019年第14期254-259,共6页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61863033);自治区自然科学基金(No.2016D01C032);自治区研究生科研创新项目(No.XJGRI2017027)

摘　　要：针对无法获得可靠羽流流向信息不利于实现羽流追踪的问题,提出了一种基于决策树的羽流追踪移动机器人自主决策方法。该方法通过移动机器人两侧的浓度传感器采集到的浓度信息,利用追踪的行为规则建立决策树模型,获得行为决策信息,使机器人高效地追踪到羽流并精确地定位。由于浓度变化关系蕴含了羽流的流向及流速信息,从而取代了传统方法中流向及流速传感器。在扩散环境下,通过移动机器人羽流追踪实验,实现了良好的源定位效果。Aiming at the problem that the reliable plume flowing direction information can not acquire, which is not con- ducive to achieve plume tracking, an autonomous decision-making method is proposed that is based on decision tree to solve the plume tracking problem of the mobile robot. In order to obtain behavioral decision information by this method, a decision tree model is established based on the concentration information collected by the concentration sensors on both sides of the robot and the tracked behavior rule, so that the robot can efficiently track the plume and accurately position it. The flow direction and velocity information of plumes are contained in the concentration relation, which replaces the flow direction and velocity sensor in the traditional method. In a diffusion environment, a good source localization effect is achieved by the mobile robot’s plume tracking experiments.

关 键 词：主动嗅觉 决策树 羽流源定位 移动机器人 羽流追踪 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]