Localization and Mapping through Multi-Sensor Fusion for Pipe Inspection: From Theory To Deployment

Abstract

Pipelines are critical infrastructure for transporting a variety of materials, such as natural gas and stormwater. However, much of this infrastructure is aging, poorly documented, and difficult to inspect, particularly pipelines that are small in diameter or buried underground. Inadequate monitoring can lead to failures with serious consequences, including service disruptions, environmental damage, and public safety risks. This thesis investigates how accurate localization and mapping can be achieved inside space-constrained, long pipelines through multi-sensor fusion, advancing both algorithmic methods and practical deployment strategies. It introduces VILL-SLAM, a system and method that combines a monocular camera, inertial measurement unit (IMU), ring-shaped laser profilometer, and LiDAR to enable real-time localization and dense RGB-D mapping with sub-millimeter reconstruction accuracy and less than 1% drift in 12-inch pipes. This approach was validated through real-world deployments in underground pipes, demonstrating its practical effectiveness. Recognizing the importance of real-world readiness, the thesis also introduces a systems engineering methodology, Design for Deployment (D4D), which emphasizes formal operational modeling, early user engagement, and iterative development through field trials. Together, these contributions offer a comprehensive, field-validated framework for in-pipe robotic inspection, advancing confined-space localization and mapping, and providing a blueprint for translating robotics research into deployable solutions for the infrastructure maintenance industry.