A Comprehensive Review of Navigation and Path-Planning Techniques for Autonomous Mobile Robots in Indoor Manufacturing Environments

Abstract

The development of Autonomous Mobile Robots (AMRs) and Automated Guided Vehicles (AGVs) has caused a major shift in the way that the manufacturing industry works, especially in terms of how material flows through the factory environment, a key signature of industry 4.0 where a manufacturing environment is required to provide flexible and intelligent material movement; nevertheless, reliable navigation and planning continue to be obstacles due to the dynamic obstacles, complexity of geometries, and limitation of sensor availability. The navigation paradigms of contemporary AMRs are reviewed based on the three main existing paradigms of navigation with respect to contemporary AMRs: sensor-based, vision-based localization and SLAM (Simultaneous Localization and Mapping)-based; this paper also identifies the key planning methods currently being applied in industrial applications. In order to successfully integrate next-generation AMRs into the industrial application, a number of hybrid frameworks have been developed that have a global and local planning layer, multimodal sensor fusion, and learning-based decision making, thus creating an alternate to the limitations presented by singular method approaches. The multi-robot coordination, conflict resolution, communication-aware planning and human-robot collaboration within smart factories are also discussed. Despite the significant advancements made in the integration of AMRs into manufacturing applications, there remain significant challenges to achieving real-time navigation in uncertain environments, enabling fully autonomous, long duration operations, providing AI transparency for safety and effectively scaling multi-robot coordination. This paper provides researchers and practitioners a framework to develop the next generation of resilient and intelligent AMR navigation solutions for manufacturing environments.