基于AprilTag与里程计融合的多智能体自定位方法  

Multi agent self positioning system based on AprilTag and odometer fusion

作　　者：孟凡宸 曹乐 王金玮 张慕谦 田健鹏 MENG Fanchen;CAO Le;WANG Jinwei;ZHANG Muqian;TIAN Jianpeng(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620

出　　处：《智能计算机与应用》2024年第7期165-169,共5页Intelligent Computer and Applications

基　　金：国家自然科学基金(61703270)。

摘　　要：针对无GNSS信号的复杂环境下的多智能体定位问题,本文提出一种基于AprilTag与里程计融合的多智能体自定位方法。首先,基于AprilTag的定位原理,实现AprilTag在ROS中的识别与测距;其次,将单目视觉定位与里程计定位相结合,经过坐标转换和坐标修正实现多智能体融合定位;最后,设计融合定位算法架构,实现定位坐标的融合计算,并在Gazebo虚拟环境中进行了定位实验。实验结果表明,该方法能仅靠摄像头和里程计实现稳定可靠的多智能体自定位。Aiming at the multi-intelligent body localization problem in the complex environment without GNSS signal,this paper proposes a multi-intelligent body self-localization method based on the fusion of AprilTag and odometer.Firstly,based on the positioning principle of AprilTag,the recognition and ranging of AprilTag in ROS are realized;secondly,monocular visual positioning is combined with odometer positioning,and multi-intelligent body fusion positioning is realized after coordinate conversion and coordinate correction;finally,the fusion positioning algorithm architecture is designed to realize the fusion calculation of positioning coordinates,and the positioning experiments are carried out in the Gazebo virtual environment.The experimental results show that the method can realize stable and reliable multi-intelligent body self-localization by only relying on the camera and odometer.

关 键 词：多智能体 Apriltag 里程计 自定位 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]