Road Graph Generator: Mapping roads at construction sites from GPS data

2024-02-15

Knowledge-Based Systems

AI Summary

Construction machinery GPS trajectories exhibit irregularity, high noise, and non-road-aligned characteristics, rendering conventional road network mapping methods—designed for standard vehicular traffic—ineffective in construction sites. Method: This paper proposes the first automated road network construction method specifically tailored to construction sites. It introduces an “anchor-driven” framework that leverages intersections as topological anchors, jointly optimizing geometric clustering and topological consistency for robust intersection detection; road segments are inferred via trajectory density modeling and graph connectivity optimization. Contribution/Results: Unlike traditional approaches reliant on canonical traffic patterns, our method is explicitly designed for construction environments. Evaluated on four real-world Norwegian construction sites, it achieves 100% accuracy in intersection and road detection under zero- or low-noise conditions, and maintains practical robustness even under high GPS noise and signal dropout. The generated road graphs effectively support construction path planning and task allocation.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Localization, Mapping, and Navigation
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

We propose a new method for inferring roads from GPS trajectories to map construction sites. This task presents a unique challenge due to the erratic and non-standard movement patterns of construction machinery, which significantly diverge from typical vehicular traffic on established roads. Our proposed method first identifies intersections in the road network that serve as critical decision points, and then connects them with edges to produce a graph, which can subsequently be used for planning and task-allocation. We demonstrate the approach by mapping roads at a real-life construction site in Norway. The method is validated on four increasingly complex segments of the map. In our tests, the method achieved perfect accuracy in detecting intersections and inferring roads in data with no or low noise, while its performance was reduced in areas with significant noise and consistently missing GPS updates.

Problem

• Inferring roads from GPS data at construction sites
• Handling erratic movement patterns of construction machinery
• Validating method accuracy in noisy GPS conditions

Innovation

• Infer roads from GPS trajectories
• Identify intersections as critical points
• Connect intersections to form road graph