基于多维全局特征融合的移动机器人地形识别
    点此下载全文
引用本文:陈青,刘晓东,周寒,邱新媛,姚建锋,刘易斯,鄢家鑫,郭星.基于多维全局特征融合的移动机器人地形识别[J].计算技术与自动化,2023,(2):20-24
摘要点击次数: 184
全文下载次数: 0
作者单位
陈青,刘晓东,周寒,邱新媛,姚建锋,刘易斯,鄢家鑫,郭星 (中国核动力研究设计院四川 成都 610213) 
中文摘要:为提高移动机器人对复杂作业场景的适应性,提出了一种基于多维全局特征融合的移动机器人地形识别方法。利用IMU传感器获取四轮移动机器人在不同地形的运动信息,采用哈希编码算法提取六通道加速度和角速度信息的全局特征,结合随机森林分类器对水泥、木板、砖石、草地和砂石路面地形进行了识别。实验结果表明,本方法可提高地形识别准确率,实现对五种地形环境的识别。
中文关键词:移动机器人  地形识别  全局特征  数据融合
 
Research on Terrain Recognition of Mobile Robot Based on Multidimensional Global Feature Fusion
Abstract:In order to improve the adaptability of mobile robot to complex operating scenes, the paper proposes a method of research on terrain recognition of mobile robot based on multidimensional global feature fusion. In the method, a IMU sensor is used to obtain the motion information of the four-wheeled robot in different terrains, and the global features of acceleration and angular velocity information of six channels are extracted by hash coding algorithm. Combined with random forest classifier, five kinds of terrains are recognized including cement pavement, wooden pavement, tile and stones pavement, grassland pavement and gravel pavement. Through experimental verification, this method can improve the accuracy of terrain recognition, identify five kinds of terrains.
keywords:mobile robot  terrain recognition  global feature  data fusion
查看全文   查看/发表评论   下载pdf阅读器