Reproduce “Vehicle Localization in GPS-Denied Scenarios Using Arc-Length-Based Map Matching” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Nur Uddin Javed, Yuvraj Singh, Qadeer Ahmed

Researchers at The Ohio State University developed an arc-length-based map matching technique to correct cumulative drift in kinematic dead reckoning for vehicle localization in GPS-denied environments. The method, leveraging 2D digital map data, improved localization accuracy with RMSE reductions often exceeding 90% across various urban driving scenarios.

Key insights

  • Static 2D map information can be effectively utilized for continuous state correction by introducing a "temporal notion" through arc-length matching with kinematic predictions.
  • Periodic re-initialization of the dead reckoning system with map-matched positions effectively mitigates cumulative error drift without relying on dynamic external measurements.
  • A self-contained localization system for GPS-denied areas can be built using standard vehicle sensors and widely available 2D maps, reducing reliance on complex or costly external infrastructure.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.