Low Cost, High Efficiency: LiDAR Place Recognition in Vineyards with Matryoshka Representation Learning

2026-01-26

AI Summary

This work addresses the challenge of efficient and accurate place recognition for mobile robots operating in structurally sparse vineyard environments that lack distinctive landmarks. To this end, the authors propose MinkUNeXt-VINE, a method that integrates a lightweight Minkowski Engine architecture, LiDAR point cloud preprocessing, and a multi-loss mechanism based on Matryoshka representation learning to achieve high accuracy and computational efficiency under low-dimensional output constraints. The approach demonstrates significantly enhanced robustness to low-cost, sparse LiDAR data and outperforms existing methods on two long-term vineyard datasets, thereby validating its effectiveness and stability in real-time applications with low-resolution sensor inputs.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Learning & Optimization for CV
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Localization in agricultural environments is challenging due to their unstructured nature and lack of distinctive landmarks. Although agricultural settings have been studied in the context of object classification and segmentation, the place recognition task for mobile robots is not trivial in the current state of the art. In this study, we propose MinkUNeXt-VINE, a lightweight, deep-learning-based method that surpasses state-of-the-art methods in vineyard environments thanks to its pre-processing and Matryoshka Representation Learning multi-loss approach. Our method prioritizes enhanced performance with low-cost, sparse LiDAR inputs and lower-dimensionality outputs to ensure high efficiency in real-time scenarios. Additionally, we present a comprehensive ablation study of the results on various evaluation cases and two extensive long-term vineyard datasets employing different LiDAR sensors. The results demonstrate the efficiency of the trade-off output produced by this approach, as well as its robust performance on low-cost and low-resolution input data. The code is publicly available for reproduction.

Problem

• place recognition
• agricultural environments
• LiDAR
• mobile robots
• unstructured environments

Innovation

• Matryoshka Representation Learning
• LiDAR Place Recognition
• MinkUNeXt-VINE
• low-cost sensing
• agricultural robotics