Skip to main navigation Skip to search Skip to main content

面向视觉惯导的导航系统初始化技术综述

Translated title of the contribution: Review of Initialization Technology for Visual-Inertial Navigation Systems
  • Yue Liu
  • , Huayi Li
  • , Shijie Zhang
  • , Chao Zhang
  • , Xiangtian Zhao
  • Harbin Institute of Technology
  • Beijing University of Posts and Telecommunications

Research output: Contribution to journalArticlepeer-review

Abstract

The visual-inertial systems are one of the most widely used basic sensor systems in navigation, and its separate use or combined use with other systems is the main scheme of sensor configuration at present. Initialization technology, as the front end of the navigation processes, determines the success or failure of navigation, so it has been studied extensively. At present, the methods related to initialization are rich in variety and engineering, but there is a lack of more comprehensive and systematic reviews. Meanwhile, although there are mature initialization paradigms, the initialization process still has a high failure rate in practice, and some initialization methods that demonstrate good navigation effects are not publicly available. With the escalation of missions and the innovation in technology, new visual-inertial systems and artificial intelligence have also brought new enhancements and challenges to the initialization. This paper summarizes the visual inertial navigation system and initialization technology, and then researches and discusses the key technologies in the initialization process in detail. After presenting classical initialization frameworks, the paper concludes with a summary and prospects.

Translated title of the contributionReview of Initialization Technology for Visual-Inertial Navigation Systems
Original languageChinese (Traditional)
Pages (from-to)1-18
Number of pages18
JournalComputer Engineering and Applications
Volume61
Issue number2
DOIs
StatePublished - 15 Jan 2025

Fingerprint

Dive into the research topics of 'Review of Initialization Technology for Visual-Inertial Navigation Systems'. Together they form a unique fingerprint.

Cite this