多移动机器人自主导航路标准确识别方法仿真  

Simulation of Autonomous Navigation Signage Accurate Recognition Method for Multiple Mobile Robots

作　　者：李晶[1] 陈思[1] LI Jing;CHEN Si(College Humanities & Information, Changchun University of Technology, Changchun Jilin 130021, China)

机构地区：[1]长春工业大学人文信息学院

出　　处：《计算机仿真》2019年第8期305-308,314,共5页Computer Simulation

摘　　要：对多移动机器人导航路标进行识别时,能有效提高识别的效率,减小机器人导航路标识别的误差,满足当前机器人导航路标识别的需求。但利用当前识别方法对机器人导航路标进行识别时,由于导航路标中干扰因素较多,降低了路标识别的准确性。提出基于角点聚类的多移动机器人导航路标识别方法。对机器人导航路标图像进行预处理,利用图像颜色分量对路标图像彩色空间进行转换,依据 Canny边缘算子定义路标图像准则函数,并对导航路标图像边缘特征检测。依据角点聚类方法对导航路标局部特征进行提取,并提出不随路标距离以及路标角度变化的路标图像局部特征表示以及匹配方法,完成对多移动机器人自主导航路标的准确识别。实验结果表明,所提方法能有效提升导航路标识别的效率与准确性。In this article, a method to recognize the multi-mobile robot navigation signpost based on corner-point clustering was proposed. Firstly, the image of robot navigation signpost was preprocessed and the color components were used to transform the color space of signpost image. Secondly, Canny edge operator was used to define the criterion function of signpost image and detect edge features of navigation signpost image. According to the corner point clustering method, the local features of navigation signpost were extracted. Meanwhile, the local feature representation and the matching method of road sign image which did not change with the signpost distance and signpost angle were put forward. Finally, the accurate recognition of autonomous navigation signpost of multi-mobile robot was completed. Simulation results show that the proposed method can effectively improve the efficiency and accuracy of navigation road sign recognition.

关 键 词：多移动机器人 自主导航 路标识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]