Joanna Szlapczynska

The reliability of positioning, navigation, and timing (PNT) data is crucial for autonomous maritime navigation, as it directly influences both the quality of navigational outcomes and overall safety. This paper presents a conceptual framework that provides data fusion based on artificial intelligence (AI) methods. The framework aims to improve the accuracy of positioning and manage the uncertainty of PNT data in the maritime environment. The study seeks to address the existing gaps in research related to data accuracy by exploring machine learning (ML) methods to support reliable autonomous shipping. These methods are expected to improve measurement precision, reduce data inaccuracies and interference, and enhance the overall robustness of positioning systems. The paper will focus on identifying the most effective AI techniques for this purpose, understanding how these methods can be implemented in various maritime scenarios, and exploring their potential to mitigate errors and enhance signal integrity. The proposed AI-based methods will be designed to remain effective in diverse scenarios, including cases of measurement absence or inaccuracy, and provide basic protection against signal attacks. The outcome of this research is a conceptual framework for a PNT system using ML techniques for Maritime Autonomous Surface Ships (MASS) with future research directions.

The literature on Microservices Architecture (MSA) outlines a range of design blueprints as well as certain detrimental practices, reflecting the diverse architectural considerations inherent in MSA design. However, it remains unclear whether and to what extent the practitioners actually adopt the good practices. The study aimed to explore how MSA practitioners apply established patterns and how they address various architectural drivers. The advantages and disadvantages of these approaches were also examined. To achieve this we conducted a survey on patterns in microservice design among a group of 77 MSA practitioners from IT companies worldwide. The survey shows a need for more accessible and standardised MSA solutions supporting MSA design phase.

Maritime transport remains integral to the global economy, facilitating the cost-efficient and scalable movement of cargo and individuals over varying distances. Modern and effective ship routing solutions not only minimize voyage time and operational costs (including fuel consumption), but also improve resource allocation and environmental sustainability. Their planning process relies heavily on optimization algorithms capable of addressing numerous environmental and operational constraints, particularly in the context of dynamic and often unpredictable weather conditions. A widely adopted approach in the literature is to formulate the ship route optimization problem as a multi-objective optimization (MOO) task, incorporating both static and dynamic constraints. The complexity of this formulation increases significantly when uncertainties related to weather conditions and ship behaviour are introduced, further complicating the optimization process. Meta-heuristic algorithms have gained prominence as effective tools for addressing i.e. complex multi-objective, constrained and nonlinear problems. Despite their demonstrated computational efficiency, the overall process of ship weather route optimization often remains computationally intensive, posing significant challenges for real-time or near-real-time applications in operational maritime contexts. High Performance Computing (HPC) emerges as a viable approach to overcome this limitation. HPC refers mostly to the use of advanced computational systems composed of parallel processing architectures (such as CUDA, OpenMP, MPI, among others) to solve much faster and more efficiently complex and data-intensive problems. Originating in scientific research domains, HPC technologies have rapidly evolved and are now being applied to support solving a wide range of problem in computer science and engineering. Employing HPC-enabled computations allows for designing a scalable and efficient framework for tackling the growing complexity of ship weather routing. This paper discusses the possibilities of HPC integration with MOO for ship weather routing, aiming to demonstrate how HPC-enabled methodologies can improve the performance and real-life applicability of the routing systems.