基于改进M-ORB的视觉SLAM直接-闭环检测算法

doi:10.11959/j.issn.2096-6652.202148

智能科学与技术学报 ›› 2021, Vol. 3 ›› Issue (4): 482-491.doi: 10.11959/j.issn.2096-6652.202148

基于改进M-ORB的视觉SLAM直接-闭环检测算法

李伟¹^,², 任孟瀚¹^,², 黄威豪¹^,³, 杜晓玉¹^,², 周毅¹^,³

¹ 河南大学计算机与信息工程学院，河南开封475004
² 河南省车联网协同技术国际联合实验室，河南开封 475004
³ 鹰驾科技（深圳）有限公司，广东深圳518052

修回日期:2021-02-26 出版日期:2021-12-15 发布日期:2021-12-01
作者简介:李伟（1979- ），女，河南大学计算机与信息工程学院副教授，主要研究方向为机器人导航、运动规划和控制
任孟瀚（1997- ），男，河南大学计算机与信息工程学院硕士生，主要研究方向为机器人导航控制、路径规划
黄威豪（1996- ），男，河南大学计算机与信息工程学院硕士生，主要研究方向为机器人导航控制、图像处理
杜晓玉（1979- ），女，博士，河南大学计算机与信息工程学院副教授，主要研究方向为无线传感器网络定位及覆盖技术
周毅（1981- ），男，博士，河南大学计算机与信息工程学院教授、博士生导师，河南省车联网协同技术国际联合实验室主任，主要研究方向为机器人导航控制、车联网技术
基金资助:
国家自然科学基金资助项目(61701170);河南省科技发展计划项目(202102210327);河南省科技发展计划项目(202102310198);河南省科技发展计划项目(202102210412)

Improved M-ORB based direct-loop closure detection algorithm for visual SLAM

Wei LI¹^,², Menghan REN¹^,², Weihao HUANG¹^,³, Xiaoyu DU¹^,², Yi ZHOU¹^,³

¹ School of Computer and Information Engineering, Henan University, Kaifeng 475004, China
² International Joint Research Laboratory for Cooperative Vehicular Networks of Henan, Kaifeng 475004, China
³ Eagle Drive Technology (Shenzhen) Co., Ltd., Shenzhen 518052, China

Revised:2021-02-26 Online:2021-12-15 Published:2021-12-01
Supported by:
The National Natural Science Foundation of China(61701170);The Program for Science and Technology Development of Henan Province(202102210327);The Program for Science and Technology Development of Henan Province(202102310198);The Program for Science and Technology Development of Henan Province(202102210412)

摘要/Abstract

摘要：

直接法SLAM不在前端提取图像特征点，使得后端无法生成视觉词袋，这导致大部分直接法SLAM无法使用带有词袋模型的闭环检测来消除系统的累积误差。针对此问题，提出一种基于改进M-ORB的视觉SLAM直接-闭环检测算法，生成闭环检测所需的词袋模型，然后采用词频-逆文档频率算法对视觉词典树各个子节点中的视觉单词进行自适应分配权重，得到场景信息的准确表述。在TUM、KITTI两种公开数据集上进行了对比实验，实验结果表明，所提出的算法能够有效检测到闭环，并在不降低准确性的同时，提高SLAM的实时性与鲁棒性。

关键词: 视觉SLAM, 闭环检测, 词袋模型, 词频-逆文档频率

Abstract:

Most kinds of direct methods do not extract image feature points in the front end of SLAM system, resulting in that they cannot use loop closure detection with bag-of-words models to eliminate the cumulative error of the system.To resolve this problem, an improved mature-oriented fast and rotated BRIEF (M-ORB) based direct-loop closure detection algorithm for visual SLAM was proposed, which designed an improved M-ORB, generated the bag of words model required for loop closure detection, and then used the term frequency-inverse document frequency (TF-IDF) algorithm to adaptively assign weights to the visual words in each sub-node of the dictionary tree.Finally, an accurate representation of the scene information was obtained.In the end, the proposed algorithm and conducted comparative experiments were verified though two public data sets TUM and KITTI.The experimental results show that the algorithm proposed in this paper can effectively detect the loop closure, and has better real-time and robustness performance without reducing the accuracy.

Key words: visual SLAM, loop closure detection, bag of words model, term frequency-inverse document frequency

中图分类号:

TP242

李伟, 任孟瀚, 黄威豪, 等. 基于改进M-ORB的视觉SLAM直接-闭环检测算法[J]. 智能科学与技术学报, 2021, 3(4): 482-491.

Wei LI, Menghan REN, Weihao HUANG, et al. Improved M-ORB based direct-loop closure detection algorithm for visual SLAM[J]. Chinese Journal of Intelligent Science and Technology, 2021, 3(4): 482-491.

图/表 17

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

表1

表2

表3

图13

图14

参考文献 21

[1]	郑南宁 . 人工智能新时代[J]. 智能科学与技术学报, 2019,1(1): 1-3.
	ZHENG N N . The new era of artificial intelligence[J]. Chinese Journal of Intelligent Science and Technology, 2019,1(1): 1-3.
[2]	刘强, 段富海, 桑勇 ,等. 复杂环境下视觉SLAM闭环检测方法综述[J]. 机器人, 2019,41(1): 112-123,136.
	LIU Q , DUAN F H , SANG Y ,et al. A survey of loop-closure detection method of visual slam in complex environments[J]. Robot, 2019,41(1): 112-123,136.
[3]	HESS W , KOHLER D , RAPP H H . Systems and methods of detecting loop closure in simultaneous localization and mapping (SLAM) applications:U.S.Patent Application,10318659B1[P]. 2019-6-11.
[4]	KLOKOV R , LEMPITSKY V . Escape from cells:deep Kd-networks for the recognition of 3D point cloud models[C]// Proceedings of 2017 IEEE International Conference on Computer Vision. Piscataway:IEEE Press, 2017: 863-872.
[5]	CHEN Z T , LAM O , JACOBSON A ,et al. Convolutional Neural Network-based Place Recognition[C]// Proceedings of the 16th Australasian Conference on Robotics and Automation.[S.l.:s.n.], 2018.
[6]	BAI D D , WANG C Q , ZHANG B ,et al. Sequence searching with CNN features for Robust and fast visual place recognition[J]. Computers and Graphics, 2018,70: 270-280.
[7]	GOMEZ-OJEDA R , LOPEZ-ANTEQUERA M , PETKOV N ,et al. Training a convolutional neural network for appearance-invariant place recognition[J]. Pattern Recognition Letters, 2015.
[8]	CHEN Z T , JACOBSON A , SUNDERHAUF N ,et al. Deep learning features at scale for visual place recognition[C]// Proceedings of 2017 IEEE International Conference on Robotics and Automation. Piscataway:IEEE Press, 2017: 3223-3230.
[9]	MERRILL N , HUANG G Q . Lightweight unsupervised deep loop closure[C]// Proceedings of Robotics:Science and Systems.[S.l.:s.n.], 2018.
[10]	SIVIC J , ZISSERMAN A . Video google:a text retrieval approach to object matching in videos[C]// Proceedings of the 9th IEEE International Conference on Computer Vision. Piscataway:IEEE Press, 2003.
[11]	CUMMINS M , NEWMAN P . Probabilistic appearance based navigation and loop closing[C]// Proceedings of 2007 IEEE International Conference on Robotics and Automation. Piscataway:IEEE Press, 2007: 2042-2048.
[12]	WILLIAMS B P , CUMMINS M , JOSé NEIRA ,et al. An image-to-map loop closing method for monocular slam[C]// Proceedings of 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway:IEEE Press, 2006: 2053-2059.
[13]	MATHIEU L , MICHAUD F . Memory management for real-time appearance-based loop closure detection[C]// Proceedings of 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway:IEEE Press, 2011: 1271-1276.
[14]	李博, 杨丹, 邓林 . 移动机器人闭环检测的视觉字典树金字塔TF-IDF得分匹配方法[J]. 自动化学报, 2011,37(6): 665-673.
	LI B , YANG D , DENG L . Visual vocabulary tree with pyramid TF-IDF scoring match scheme for loop closure detection[J]. Acta Automatica Sinica, 2011,37(6): 665-673.
[15]	GAO X , WANG R , DEMMEL N ,et al. LDSO:direct sparse odometry with loop closure[C]// Proceedings of 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway:IEEE Press, 2018: 2198-2204.
[16]	RUBLEE E , RABAUD V , KONOLIGE K ,et al. ORB:an efficient alternative to SIFT or SURF[C]// Proceedings of the 13th IEEE International Conference on Computer Vision. Piscataway:IEEE Press, 2011: 2564-2571.
[17]	ENGEL J , USENKO V , CREMERS D . A photometrically calibrated benchmark for monocular visual odometry[J]. arXiv preprint,2016,arXiv:1607.02555.
[18]	GEIGER A , LENZ P , URTASUN R . Are we ready for autonomous driving? The KITTI vision benchmark suite[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE Press, 2012: 3354-3361.
[19]	MUR-ARTAL R , TARDOS J D . ORB-SLAM2:an open-source SLAM system for monocular,stereo and RGB-D cameras[J]. IEEE Transactions on Robotics, 2017,33(5): 1255-1262.
[20]	ENGEL J , SCH T , CREMERS D . LSD-SLAM:large-scale direct monocular SLAM[C]// Proceedings of European Conference on Computer Vision.[S.l.:s.n.], 2014: 834-849.
[21]	ENGEL J , KOLTUNV , CREMERSD . Direct sparse odometry[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018,40(3): 611-625.

KITTI序列	DSO	ORB-SLAM2	D-SLAM
KITTI_00	0.298 4%	0.246 6%	0.215 3%
KITTI_05	0.187 7%	0.151 3%	0.114 9%
KITTI_07	0.287 4%	0.289 8%	0.159 2%
KITTI_09	0.216 4%	0.183 9%	0.164 9%

KITTI序列	DSO	ORB-SLAM2	D-SLAM
KITTI_00	48.278 1	1.297 5	0.621 7
KITTI_05	41.519 7	0.951 7	0.607 4
KITTI_07	40.476 2	0.769 1	0.445 7
KITTI_09	52.175 9	1.459 2	0.683 7

KITTI序列	DSO	ORB-SLAM2	D-SLAM
KITTI_00	2.981 6	1.459 2	0.627 4
KITTI_05	2.274 3	1.297 5	0.601 7
KITTI_07	1.671 4	0.951 7	0.445 7
KITTI_09	2.999 4	1.679 1	0.694 2

基于改进M-ORB的视觉SLAM直接-闭环检测算法

Improved M-ORB based direct-loop closure detection algorithm for visual SLAM

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献 21

相关文章 15

Metrics

推荐阅读 0

[1]	张子菁, 章飞. 基于位姿图优化的势均衡多伯努利滤波器SLAM方法[J]. 智能科学与技术学报, 2023, 5(1): 113-120.
[2]	曹翔, 孙长银. 基于不确定事件威胁度评估的UUV任务重规划[J]. 智能科学与技术学报, 2022, 4(4): 493-502.
[3]	张佳欣, 张森林, 刘妹琴, 董山玲, 郑荣濠. 面向海洋环境自适应采样的多AUV协同定位[J]. 智能科学与技术学报, 2022, 4(4): 503-512.
[4]	许一航, 刘剑, 孙长银. 基于四元数的全驱动碟形AUV单矢量反馈控制[J]. 智能科学与技术学报, 2022, 4(4): 513-521.
[5]	蔡莹皓, 杨华, 安璇, 王文硕, 杜沂东, 张嘉韬, 王志刚. 神经符号学及其应用研究[J]. 智能科学与技术学报, 2022, 4(4): 560-570.
[6]	赵超, 许杰, 陈星宇, 梅魁志, 兰旭光. 机器人持续学习进展与展望[J]. 智能科学与技术学报, 2022, 4(3): 308-323.
[7]	崔少伟, 王硕, 胡静怡, 张超凡. 面向机器人操作任务的视触觉传感技术综述[J]. 智能科学与技术学报, 2022, 4(2): 186-199.
[8]	徐德, 秦方博. 机器人自动轴孔装配研究进展[J]. 智能科学与技术学报, 2022, 4(2): 200-211.
[9]	葛悦光, 张少林, 蔡莹皓, 鲁涛, 温大勇, 王海涛, 王硕. 本体知识表示方法在机器人领域的应用研究综述[J]. 智能科学与技术学报, 2022, 4(2): 212-222.
[10]	胡静怡, 崔少伟, 张超凡, 张伯约, 王硕. 基于触觉感知和伺服的物体三维边缘重建方法[J]. 智能科学与技术学报, 2022, 4(2): 233-245.
[11]	董璐, 熊爱玲. 基于改进RRT*-Smart的复杂动态环境下的无人艇路径规划[J]. 智能科学与技术学报, 2022, 4(2): 264-276.
[12]	沈继锋, 刘岳, 韦浩, 左欣, 杨万扣. 基于差分特征注意力机制的无锚框多光谱行人检测算法[J]. 智能科学与技术学报, 2021, 3(3): 294-303.
[13]	王腾, 潘晶, 董璐, 孙长银. 面向防疫的智能导诊机器人关键技术及应用[J]. 智能科学与技术学报, 2021, 3(2): 187-194.
[14]	邓晓燕, 俞祝良, 林灿光, 顾正晖, 李远清. 基于脑机接口的人机共享控制技术研究[J]. 智能科学与技术学报, 2021, 3(1): 85-92.
[15]	王日中, 李慧平, 崔迪, 徐德民. 基于深度强化学习算法的自主式水下航行器深度控制[J]. 智能科学与技术学报, 2020, 2(4): 354-360.