Abstract
Achieving robust cognitive autonomy in robots navigating complex,unpredictable environments remains a fundamental challenge in robotics. Thispaper presents Underwater Robot Self-Organizing Autonomy (UROSA), agroundbreaking architecture leveraging distributed Large Language Model AIagents integrated within the Robot Operating System 2 (ROS 2) framework toenable advanced cognitive capabilities in Autonomous Underwater Vehicles. UROSAdecentralises cognition into specialised AI agents responsible for multimodalperception, adaptive reasoning, dynamic mission planning, and real-timedecision-making. Central innovations include flexible agents dynamicallyadapting their roles, retrieval-augmented generation utilising vector databasesfor efficient knowledge management, reinforcement learning-driven behaviouraloptimisation, and autonomous on-the-fly ROS 2 node generation for runtimefunctional extensibility. Extensive empirical validation demonstrates UROSA'spromising adaptability and reliability through realistic underwater missions insimulation and real-world deployments, showing significant advantages overtraditional rule-based architectures in handling unforeseen scenarios,environmental uncertainties, and novel mission objectives. This work not onlyadvances underwater autonomy but also establishes a scalable, safe, andversatile cognitive robotics framework capable of generalising to a diversearray of real-world applications.