Skip to main navigation Skip to search Skip to main content

An RGB-Thermal-Inertial Odometry with Dual-Track Front-end and Asynchronous Factor Graph Optimization

  • Pei Wang
  • , Bo Liu
  • , Qinghua Li*
  • , Siyuan Wang
  • *Corresponding author for this work
  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The localization of unmanned systems in degraded visual environments (DVEs) has emerged as a critical research challenge in recent years. This paper studies the issues of robust multi-modal fusion and sensor degradation for RGB-Thermal-Inertial Odometry frameworks. A dual-track front-end method combining deep-learning descriptors and KLT optical flow is designed, ensuring reliable cross-modal association and efficient temporal tracking. Meanwhile, a scene-aware feature adapter and an asynchronous factor graph are developed to address sensor anomalies through dynamic sensor balancing and seamless monocular-binocular transitions. The effectiveness and superiority of the proposed framework are validated through real-world experiments in lighting transitions and smoke scenarios, significantly outperforming state-of-the-art systems with high localization accuracy at real-time frequencies.

Original languageEnglish
Title of host publicationProceedings of ISoIRS 2026 - Moving Towards Embodied Intelligence in the AI Age
Subtitle of host publication2026 6th International Symposium on Intelligent Robotics and Systems
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798319520029
DOIs
StatePublished - 2026
Event6th International Symposium on Intelligent Robotics and Systems, ISoIRS 2026 - Shenzhen, China
Duration: 27 Mar 202629 Mar 2026

Publication series

NameProceedings of ISoIRS 2026 - Moving Towards Embodied Intelligence in the AI Age: 2026 6th International Symposium on Intelligent Robotics and Systems

Conference

Conference6th International Symposium on Intelligent Robotics and Systems, ISoIRS 2026
Country/TerritoryChina
CityShenzhen
Period27/03/2629/03/26

Keywords

  • Odometry
  • State estimator
  • Thermal
  • Visual
  • scene-aware feature adapter

Fingerprint

Dive into the research topics of 'An RGB-Thermal-Inertial Odometry with Dual-Track Front-end and Asynchronous Factor Graph Optimization'. Together they form a unique fingerprint.

Cite this