A Survey-Driven Framework for Autonomous Mobile Robot Navigation Systems: The Perception–Cognition–Operation (PCO) Approach

This paper introduces a novel theoretical framework for classifying Autonomous Mobile Robots (AMRs) into three hierarchical layers: Perception, Cognition, and Operation. Unlike prior hardware-centric taxonomies, our approach, grounded in a structured review of seminal works, foundational methodologies, and state-of-the-art advances, explicitly integrates locomotion mechanisms (wheeled, legged), application domains (industrial, agricultural), and autonomy levels with navigation strategies. The framework unifies terrestrial navigation techniques into a cohesive taxonomy, clarifying modular boundaries and interdependencies. Serving as both a conceptual guide and educational tool, it empowers researchers to evaluate trade-offs in sensor configurations, decision-making algorithms, and trajectory execution under real-world constraints. A comparative analysis positions this framework against established navigation architectures, highlighting its role as a high-level reference design for modular implementations. By bridging theoretical principles with system optimization, the framework enhances interoperability across robotic platforms. Ultimately, this work delivers a practical design atlas, structuring the end-to-end pipeline of autonomous navigation to guide researchers and practitioners in selecting algorithms suited to their specific robotic platforms and mission requirements.