关键词: 主动嗅觉, 决策树, 羽流源定位, 移动机器人, 羽流追踪

Abstract: Aiming at the problem that the reliable plume flowing direction information can not acquire, which is not conducive 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.

Key words: active olfactory, decision tree, plume source location, mobile robot, plume stracking