Hybrid automata-based software architecture and simulation tools for mission planning of marine vehicles

Modern maritime autonomous navigation should adapt to dynamic environments and offer interpretable and certifiable safety guarantees while maintaining compliance with the International Regulations for Preventing Collisions at Sea (COLREGs).We present a hybrid automaton based framework that combines decision making capabilities while being transparent and verifiable. Our approach structures vessel behavior into interpretable modes and enables the design of motion planning algorithms based on rule-driven switching between modes. We introduce HybrautNav, a modular ROS 2-based navigation stack, packaged as a Docker-deployable module suitable for embedded platforms. We present the software architecture that integrates HybrautNav into a mission-planning system with real-time risk assessment and scenario management. We use a configurable USV simulation suite on representative encounter scenarios, measuring collision rates, COLREGs compliance metrics, and operator interpretability.