Back to research modules

Research report · Research Prototype

SHIELD-VIO: Self-Healing Visual-Inertial Odometry

Robust localization research prototype · 2026–Present

SHIELD-VIO frames localization robustness as a closed-loop autonomy problem: the robot should monitor localization health, reason about degradation causes, and select recovery actions before failure becomes safety-critical.

Abstract

Study whether visual-inertial odometry systems can detect degradation, diagnose likely failure causes, and select recovery actions before localization failure.

Scientific question: Can a VIO pipeline become health-aware enough to anticipate and mitigate localization degradation?

Contribution: A modular health-monitoring and diagnosis scaffold for robust VIO, linking estimator confidence to recovery actions.

Marked research artifact slot

Health-aware VIO architecture

Architecture figure slot for verified diagrams of modules, data flow, and software boundaries.

Synthetic / planned visual

Marked research artifact slot

Detect · diagnose · recover pipeline

Pipeline slot for GIFs or animations that explain estimation, planning, perception, or evaluation steps.

Synthetic / planned visual

Marked research artifact slot

Synthetic VIO uncertainty timeline

Demo slot reserved for real project videos or synthetic demos that are explicitly labeled as synthetic.

Synthetic / planned visual

Research questions

  • Can VIO degradation be represented as a continuous health signal before tracking loss?
  • Can onboard signals infer a useful posterior over degradation causes?
  • Can diagnosis-conditioned recovery improve estimator uptime and consistency?

Methods

Health monitoringBayesian failure diagnosisNavigation Health IndexRule-based recovery policies

Implementation status

Implemented

  • Repository-level research framing
  • Health-index concept
  • Failure-mode taxonomy
  • Recovery-policy scaffold

Research prototype

  • Synthetic degradation timeline
  • Estimator health visualization
  • Diagnosis-normalization tests

Planned

  • ROS 2 node integration
  • OpenVINS or ORB-SLAM3 interface
  • EuRoC/TUM-VI degradation benchmark

Experiments and metrics

Experiments

  • Synthetic degradation demo
  • Detector range tests
  • Diagnosis normalization tests
  • Recovery-action smoke tests

Metrics

  • Navigation Health Index
  • failure-detection latency
  • ATE/RPE after degradation
  • recovery success rate — planned

Quantitative benchmark tables will be added only after reproducible experiments are available.

Limitations

  • Quantitative claims are intentionally withheld until reproducible benchmark runs are available.
  • Visual figures and GIFs are placeholders unless a project page explicitly states that an artifact is generated from real experiments.

Reproducibility plan

Experiment workflow

Each module is prepared to document configuration files, datasets, evaluation metrics, and repeated experiment runs before quantitative claims are shown.

Repository setup

Public repositories are linked when available. Private, planned, or incomplete repositories remain clearly labelled rather than being presented as finished systems.

Literature context

  • Visual-inertial odometry
  • Robust SLAM
  • Perception-aware autonomy
  • Fault diagnosis in robotics

Roadmap

  • Integrate with OpenVINS or ORB-SLAM3
  • Add ROS 2 health monitor node
  • Evaluate on degraded EuRoC/TUM-VI sequences