Designing a reliable fiducial marker network for autonomous indoor robot navigation

Automation and robotics offer significant potential to address some of the challenges faced by facility managers to efficiently operate and maintain indoor building environments. Previous efforts have focused on deploying mobile service robots for scheduled and periodic tasks such as monitoring, inspecting, and collecting data. Localization and navigation are two of the fundamental capabilities required for any robotic system to accomplish these periodic tasks successfully. Most of the existing approaches for achieving semi/fully autonomous indoor mobile robot navigation either require dense instrumentation of the physical space (e.g., Bluetooth beacons) or are computationally burdensome (e.g., Simultaneous Localization And Mapping). To address these issues, the authors previously developed localization, navigation, and drift correction algorithms based on cost-effective and easily-reconfigurable fiducial markers (e.g., AprilTags). However, these algorithms were based on context-specific assumptions regarding the marker characteristics, sensor capabilities, and environmental conditions. This study comprehensively investigates the design characteristics of a fiducial marker network localization system to achieve autonomous mobile indoor navigation. A generalized framework in the form of a process flow chart is proposed that is agnostic of indoor building environment application, marker category, robot, and facility type. That is, the proposed framework can be used to systematically to design the desired robot, required sensors, and create the optimal marker network map. The feasibility of the proposed approach is explained with the help of a facility management related example. The outcomes of this study can be generally applicable to any mobile robot, building type (e.g., office), and application (e.g., construction progress monitoring).