PointNetPGAP-SLC: A 3D LiDAR-based Place Recognition Approach with Segment-level Consistency Training for Mobile Robots in Horticulture

TL;DR

This paper introduces PointNetPGAP, a novel LiDAR-based place recognition model with segment-level consistency training, specifically designed for horticultural environments, improving robustness and accuracy in challenging semi-permeable and ambiguous settings.

Contribution

The paper presents a new model combining statistical aggregators, a segment-level consistency training approach, and a horticultural dataset, advancing LiDAR place recognition in horticultural environments.

Findings

PointNetPGAP outperforms existing models on horticultural datasets.

Segment-level consistency training enhances descriptor robustness.

Model achieves superior retrieval accuracy in ambiguous, semi-permeable environments.

Abstract

3D LiDAR-based place recognition remains largely underexplored in horticultural environments, which present unique challenges due to their semi-permeable nature to laser beams. This characteristic often results in highly similar LiDAR scans from adjacent rows, leading to descriptor ambiguity and, consequently, compromised retrieval performance. In this work, we address the challenges of 3D LiDAR place recognition in horticultural environments, particularly focusing on inter-row ambiguity by introducing three key contributions: (i) a novel model, PointNetPGAP, which combines the outputs of two statistically-inspired aggregators into a single descriptor; (ii) a Segment-Level Consistency (SLC) model, used exclusively during training to enhance descriptor robustness; and (iii) the HORTO-3DLM dataset, comprising LiDAR sequences from orchards and strawberry fields. Experimental evaluations conducted on the HORTO-3DLM and KITTI Odometry datasets demonstrate that PointNetPGAP outperforms state-of-the-art models, including OverlapTransformer and PointNetVLAD, particularly when the SLC model is applied. These results underscore the model's superiority, especially in horticultural environments, by significantly improving retrieval performance in segments with higher ambiguity.