Semantic-Aware Particle Filter for Reliable Vineyard Robot Localisation

2025-09-22

AI Summary

In structured outdoor environments such as vineyards, LiDAR-based localization often fails due to geometric repetition between rows and perceptual ambiguity. To address this, we propose a semantic-aware particle filter localization method. Our approach fuses LiDAR measurements with object-level semantic landmarks—specifically vine trunks and support posts—and introduces a novel “semantic wall” mechanism that constructs pseudo-structural constraints from adjacent semantic landmarks to suppress inter-row ambiguity. In semantically sparse regions, it adaptively incorporates low-accuracy GPS priors to ensure global consistency. Experimental evaluation in real-world vineyards demonstrates stable intra-row localization, robust recovery when AMCL fails, and significantly improved accuracy and robustness compared to vision-based SLAM systems such as RTAB-Map.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Computer Vision: Low Level & Physics-based Vision
• Search and Optimization: Local Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Accurate localisation is critical for mobile robots in structured outdoor environments, yet LiDAR-based methods often fail in vineyards due to repetitive row geometry and perceptual aliasing. We propose a semantic particle filter that incorporates stable object-level detections, specifically vine trunks and support poles into the likelihood estimation process. Detected landmarks are projected into a birds eye view and fused with LiDAR scans to generate semantic observations. A key innovation is the use of semantic walls, which connect adjacent landmarks into pseudo-structural constraints that mitigate row aliasing. To maintain global consistency in headland regions where semantics are sparse, we introduce a noisy GPS prior that adaptively supports the filter. Experiments in a real vineyard demonstrate that our approach maintains localisation within the correct row, recovers from deviations where AMCL fails, and outperforms vision-based SLAM methods such as RTAB-Map.

Problem

• Addresses robot localization failures in vineyards due to repetitive row geometry
• Mitigates perceptual aliasing by incorporating semantic object-level detections
• Maintains global consistency in sparse semantic regions like headland areas

Innovation

• Semantic particle filter fuses object detections with LiDAR
• Semantic walls connect landmarks to create pseudo-structural constraints
• Adaptive GPS prior maintains global consistency in sparse areas