Research Analyzer
← Back ICRA 2026

Real-Time Geometric-Registration-Based Precision Localization for Autonomous Docking in Unstructured Factory Environment

Sebastian Fernando Chinchilla Gutierrez, Manaru Watanabe, Masahiro Ooyama, Takayuki Yamada, Tomoaki Yamada, Naoto Toshiki, Satsuki Yamane, Jose Victorio Salazar Luces, Ankit A. Ravankar, Yasuhisa Hirata

PDF

AI summary

Key figure (auto-extracted from paper)
A real-time geometric registration method achieves high-precision autonomous docking in sparse, dynamic factory environments, outperforming traditional methods by significantly reducing errors and increasing success rates.
autonomous docking geometric registration sparse point clouds Hough transform factory automation real-time localization

Problem

Autonomous mobile robots struggle to dock precisely in unstructured factories due to sparse point clouds, dynamic landmarks, occlusions, and repetitive patterns that confuse standard clustering and registration algorithms.

Approach

The system uses the Hough transform to detect and describe line features in sparse 2D point clouds, filters irrelevant lines, and aligns them with a predefined reference pattern to estimate poses in real time.

Key results

  • Achieved ±5.06 mm and ±1.11° docking accuracy across 72 factory trials
  • Attained 100% docking success rate while correctly identifying target carts
  • Reduced localization errors by 70% and increased success rates by 86% versus baselines
  • Enabled real-time processing with low computational overhead for sparse data

Why it matters

This approach enables reliable, high-precision autonomous material handling in dynamic industrial settings where traditional navigation fails, advancing smart manufacturing and flexible automation.

Abstract

In factory distribution processes, autonomous mo- bile robots must dock precisely at base stations. However, this task is challenging due to the dynamic and unstructured nature of factory environments, as well as the sparse point clouds caused by sensor occlusions and distance limitations. To address these challenges, we propose a geometric registration approach designed to handle sparse point clouds in changing, unstructured settings. Our method utilizes the Hough transform to detect lines, describes the point cloud based on the relationships between these lines, filters out lines that do not correspond to the geometric features of the target base station, and estimates the pose of both the station and the robot using global registration techniques. We evaluated our system in four typical factory scenarios across 72 trials. Results show the robot achieved docking accuracy within ±5.06 mm and ±1.11°, with a 100% success rate in docking and correctly identifying the target cart from surrounding objects. This represents a 70% reduction in errors and an 86% increase in success rate compared to existing methods.

Index terms

Autonomous Vehicle Navigation Intelligent Transportation Systems Intelligent and Flexible Manufacturing

Related papers