Article Details

Title · Authors · Abstract · Keywords · DOI · References

Sparse Map Construction Based on ORB-SLAM3 Monocular Vision

HU Xiaowei, XUE Dijie, SUN Yu

(Sias University, Zhengzhou 451150, China)

Section：Computer Technology · Vol. 2025, Issue 5 · Pages：125-128,133 · 2096-4706（2025）05-0125-05 · CLC number：TP242 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2025.05.023

Abstract

This paper introduces a mapping method based on ORB-SLAM3 sparse map, which enables the robot to realize the construction of sparse map without radar. The robot mainly has two core control units of NVIDIA Jetson Nano and STM32. The former is mainly used to run the ROS system and process the camera data connected to it. The latter is mainly used to control the chassis motion and process the ultrasonic sensor and gyroscope data connected to it. The two processors communicate through a serial port. Through the barrage camera, the robot uses the ORB feature as the core of the visual SLAM, matches feature points through the key frames of different angles, realizes the visual odometry, and performs tracking and loop closure detection, and finally completes the construction of the sparse map.

Keywords

ROS; ORB; SLAM

References

  1. [1] Xu Haonan, Yu Lei, Fei Shumin Large scene dense 3D reconstruction system based on semi direct SLAM [J]. Pattern Recognition and Artificial Intelligence, 2018, 31 (5): 477-484
  2. [2] ZHOU Y，YAN F H，ZHOU Z. Handling Pure Camera Rotation in Semi-Dense Monocular SLAM [J].The Visual Computer，2019，35（1）：123-132.
  3. [3] Still in charge Implementation of a visual SLAM monocular semi dense mapping method [J]. Intelligent Computers and Applications, 2021, 11 (1): 89-93
  4. [4] YU C，LIU Z X，LIU X J，et al. DS-SLAM: A Semantic Visual SLAM towards Dynamic Environments [C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).Madrid：IEEE，2018：1168-1174.
  5. [5] Ma Jingxuan, Wang Hongyu, Cao Yan, etc Implementation of Single Object ORB-SLAM Dense Mapping for Embedded Platforms [J]. Computer Engineering and Applications, 2022, 58 (16): 213-218
  6. [6] LI L L，ZHANG R R，CHEN L P，et al. Spray Drift Evaluation with Point Clouds Data of 3D LiDAR as a Potential Alternative to the Sampling Method [J/OL].Frontiers in Plant Science，2022，13：939733[2024-09-20].https://doi.org/10.3389/fpls.2022.939733.
  7. [7] Durrant-Whyte H F. Uncertain Geometry in Robotics [J].IEEE Journal on Robotics and Automation，1988，4（1）：23-31.
  8. [8] CUPEK R，DREWNIAK M，FOJCIK M，et al. Autonomous Guided Vehicles for Smart Industries-The State-of-the-Art and Research Challenges [C]//International Conference on Computational Science.Springer，2020：330-343.
  9. [9] WOLTERS S H G. Accelerated Implementation of 3D Visual EKF SLAM For Handheld Perfusion Imaging [D].Enschede：University of Twente，2021.
  10. [10] ISLAM S，ISLAM T，FARUQUE M G G. A Survey on LiDAR-Based SLAM Technique for an Autonomous Model Using Particle Filters [C]//Proceedings of ICSCS 2021.Singapore：Springer，2022：227-240.

Funding

Henan Province Science and Technology Research Project (232102220071); Discipline and Major Construction Funding Project for Private Ordinary Higher Education Institutions in Henan Province in 2022 (Education Office Political Science and Law [2021] No. 255, Communication Engineering Major); Zhengzhou Sias College General Research Project for 2024 (2024XKD046)

About the authors

Hu Xiaowei (1980-), male, Han nationality, from Nanyang, Henan Province, associate professor, undergraduate, research direction: machine vision.