AUTHOR=Li Xiangyu S. , Nguyen T. L. , Cohn Anthony G. , Dogar Mehmet , Cohen Netta 

TITLE=Real-time robot topological localization and mapping with limited visual sampling in simulated buried pipe networks

JOURNAL=Frontiers in Robotics and AI

VOLUME=10

YEAR=2023

URL=https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2023.1202568

DOI=10.3389/frobt.2023.1202568

ISSN=2296-9144

ABSTRACT=<p><bold>Introduction:</bold> Our work introduces a real-time robotic localization and mapping system for buried pipe networks.</p><p><bold>Methods:</bold> The system integrates <italic>non-vision-based</italic> exploration and navigation with an <italic>active-vision-based</italic> localization and topological mapping algorithm. This algorithm is selectively activated at topologically key locations, such as junctions. <italic>Non-vision-based</italic> sensors are employed to detect junctions, minimizing the use of visual data and limiting the number of images taken within junctions.</p><p><bold>Results:</bold> The primary aim is to provide an accurate and efficient mapping of the pipe network while ensuring real-time performance and reduced computational requirements.</p><p><bold>Discussion:</bold> Simulation results featuring robots with fully autonomous control in a virtual pipe network environment are presented. These simulations effectively demonstrate the feasibility of our approach in principle, offering a practical solution for mapping and localization in buried pipes.</p>