Visual Inertial SLAM System Based on Embedded Parallel Processing

DOI

10.2991/amcce-18.2018.55How to use a DOI?

Keywords

Visual Inertial, Parallel Processing, Nonlinear Optimization

Abstract

Based on the embedded computing platform, we propose a visual inertial fusion SLAM system. The software algorithm combines the matching of monocular visual feature recognition and IMU measurement pre-integration algorithms. And the back-end uses tightly-integrated nonlinear optimization algorithms to process data. The hardware uses NVIDIA's Jetson TX1 parallel processing computing platform and low-cost sensors to retrieve data. The robot uses the open source robot operating system ROS as the operating system and the upper computer uses the UBUNUT system. Based on the embedded parallel processing, visual inertial SLAM system has the characteristics of low cost and good stability.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jianyue Zhang
AU  - Ge Li
AU  - Xuehe Zhang
PY  - 2018/05
DA  - 2018/05
TI  - Visual Inertial SLAM System Based on Embedded Parallel Processing
BT  - Proceedings of the 2018 3rd International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 2018)
PB  - Atlantis Press
SP  - 317
EP  - 323
SN  - 2352-5401
UR  - https://doi.org/10.2991/amcce-18.2018.55
DO  - 10.2991/amcce-18.2018.55
ID  - Zhang2018/05
ER  -