Research Analyzer
← Back ICRA 2026

XPRESS: X-Band Radar Place Recognition Via Elliptical Scan Shaping

Hyesu Jang, Wooseong Yang, Ayoung Kim, DONGJE LEE, Hanguen Kim

PDF

AI summary

Key figure (auto-extracted from paper)
XPRESS enables robust, GPS-denied maritime navigation by intentionally degrading low-resolution X-band radar scans into elliptical shapes for fast, rotationally invariant place recognition.
X-band radar place recognition maritime navigation elliptical scan shaping rotational invariance autonomous vessels

Problem

X-band radar is ubiquitous on maritime vessels but its low resolution, high noise, and susceptibility to rotational drift hinder reliable place recognition for autonomous navigation in complex coastal environments.

Approach

The method clusters radar returns, approximates them with ellipses, and converts them into polar histograms to create a rotationally invariant descriptor, while using cluster counts for rapid candidate filtering.

Key results

  • First X-band radar-specific place recognition algorithm for maritime use
  • Robust retrieval performance across public MOANA, Pohang Canal, and custom datasets
  • Fast candidate filtering via integer-based cluster count thresholding
  • Proven rotational invariance and resilience to dynamic vessel fluctuations

Why it matters

Provides a practical, hardware-compatible solution for reliable autonomous vessel navigation in GPS-denied coastal and port areas.

Abstract

X-band radar serves as the primary sensor on mar- itime vessels, however, its application in autonomous navigation has been limited due to low sensor resolution and insufficient information content. To enable X-band radar-only autonomous navigation in maritime environments, this paper proposes a place recognition algorithm specifically tailored for X-band radar, incorporating an object density-based rule for efficient candidate selection and intentional degradation of radar detections to achieve robust retrieval performance. The proposed algorithm was evaluated on both public maritime radar datasets and our own collected dataset, and its performance was compared against state-of-the-art radar place recognition methods. An ablation study was conducted to assess the algorithm’s performance sensitivity with respect to key parameters.

Index terms

Marine Robotics Range Sensing Localization

Related papers