Assurance methodology for Maritime Autonomous Surface Ships

Next-generation Maritime Autonomous Surface Ships (MASS) will rely on several cyber-physical subsystems, whereas humans' involvement is expected for their supervision and/or remote operation. MASS systems diverse characteristics during their lifecycle, namely, complexity, heterogeneity, progressivity, non-predictability, among others, raise immense challenges for their assurance.
This research aims at proposing a novel methodology for the MASS assurance developing: (a) methods to determine testing scenarios, and; (b) interconnected digital tools, lab-scale models and full-scale test sites and systems, providing pathways to MASS testing, verification, and validation. The testing scenarios determination method will be based on integrating methods and tools for statistical and big data analysis along with safety analysis and system models. Digital twins of the physics-based, data-driven and hybrid types will be developed to represent the behaviour of the case MASSs and their systems, considering the critical functions that must be retained during their lifetime. Effective algorithms for decision-making, multi sensorial data fusion as well as intelligent monitoring and diagnosis will be developed. The uncertainty of the complete chain from sensors to the decision making as well as its impact will systematically be studied.
The testing, verification, and validation methodology to facilitate the MASS assurance will follow a multi-phase approach considering testing in virtual environment, lab-scale and full-scale whilst accommodating both normal and extreme (risky) scenarios. The virtual testing platform will be based on the developed digital tools, whereas available tools like Open Simulation Platform will be considered. Interconnections of the virtual testing platform with lab-scaled model testing and full-scale testing will be proposed to effectively, efficiently, and timely fulfil the set testing requirements. The lab-scaled models will be designed to provide rapid...