Virtual Beam aided SINS/DVL Tightly Coupled Integration Method with Partial DVL Measurements

Yiqing Yao, Xiang Xu, Xiaosu Xu, Itzik Klein

Research output: Contribution to journalArticlepeer-review


Strapdown inertial navigation system (SINS)/ Doppler velocity log (DVL) tightly coupled (TC) fusion is a popular scheme as it offers increased robustness compared to other approaches. In TC, DVL beam velocities are separately integrated with SINS in the navigation filter. To enhance the performance of SINS/DVL TC integration system in situations of partial DVL beam measurements, virtual beam aided solutions are proposed, theoretically derived, and experimentally evaluated. To that end, we analyze the geometrical relation between the DVL beam configuration and a zero velocity vector assumption and derive a zero velocity update (ZUPT) aided virtual beam method to continuously operate the TC fusion. Further, to avoid the precondition limitation of ZUPT method, a least squares support vector machine (LS-SVM) aided virtual beam construction algorithm together with its SINS/DVL TC integration scheme is proposed. Our experiments show the advantages and disadvantages of the ZUPT aided solution under different scenarios, and indicate the effectiveness of the LS-SVM aided virtual beam solution during short partial beam measurement periods.

Original languageEnglish
Pages (from-to)1-9
Number of pages9
JournalIEEE Transactions on Vehicular Technology
StateAccepted/In press - 2022

Bibliographical note

Publisher Copyright:


  • Acoustic beams
  • Artificial intelligence
  • artificial intelligence
  • Navigation
  • partial DVL measurement
  • Sea measurements
  • SINS/DVL tightly coupled system
  • Structural beams
  • Vehicle dynamics
  • Velocity measurement
  • virtual beam construction
  • ZUPT

ASJC Scopus subject areas

  • Automotive Engineering
  • Aerospace Engineering
  • Electrical and Electronic Engineering
  • Applied Mathematics


Dive into the research topics of 'Virtual Beam aided SINS/DVL Tightly Coupled Integration Method with Partial DVL Measurements'. Together they form a unique fingerprint.

Cite this