首页 | 本学科首页   官方微博 | 高级检索  
     检索      

单目视觉惯性的同时定位与地图构建综述
引用本文:瑚 琦,蔡文龙,卢定凡,姜 敏.单目视觉惯性的同时定位与地图构建综述[J].教育技术导刊,2020,19(7):275-280.
作者姓名:瑚 琦  蔡文龙  卢定凡  姜 敏
作者单位:1. 上海理工大学 光电信息与计算机工程学院;2. 上海市现代光学系统重点实验室,上海 200093
基金项目:上海市科委科学仪器科技攻关项目(12142200400)
摘    要:在机器人领域,同时定位与地图构建(SLAM)是导航定位的关键技术。单目视觉传感器由于结构简单、成本低且能获取丰富的视觉信息,被广泛应用于 SLAM。随着无人机、AR 设备以及自动驾驶汽车技术的快速发展,视觉惯性 SLAM(VI-SLAM)技术得到了越来越多人的关注。针对 VI-SLAM,从滤波与非线性优化的角度出发,首先介绍相关算法理论框架,然后分析几种具有代表性的单目 VI-SLAM 算法创新点及实现方案,并通过 EuRoC MAV 数据集评估各算法优劣,最后结合深度学习与语义 SLAM,对 SLAM 未来发展趋势进行探讨。

关 键 词:视觉惯性  SLAM  滤波  非线性优化  深度学习  
收稿时间:2019-09-09

Survey on Monocular Visual Inertial SLAM Algorithms
HU Qi,CAI Wen-long,LU Ding-fan,JIANG Min.Survey on Monocular Visual Inertial SLAM Algorithms[J].Introduction of Educational Technology,2020,19(7):275-280.
Authors:HU Qi  CAI Wen-long  LU Ding-fan  JIANG Min
Institution:1.School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology; 2.Shanghai Key Lab of Modern Optical System,Shanghai 200093,China
Abstract:In the field of robotics,simultaneous localization and mapping(SLAM)is the key technology for navigation and positioning. Monocular vision sensors are widely used in SLAM due to their simple structure,low cost,and access to rich visual information. With the rapid development of drones,AR equipment and self-driving cars,visual inertial SLAM(VI-SLAM)technology has received more and more attention. For VI-SLAM,from the filtering-based and optimization-based perspective,the basic theory of algorithm implementation is introduced first. Secondly,the innovation points and implementation schemes of several representative monocular VI-SLAM algorithms are analyzed. Then,the advantages and disadvantages of each algorithm are evaluated through the EuRoC MAV dataset. Finally,combined with deep learning and semantic SLAM,the future development trend of SLAM is discussed.
Keywords:VI-SLAM  filter  nonlinear optimization  deep learning  
点击此处可从《教育技术导刊》浏览原始摘要信息
点击此处可从《教育技术导刊》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号