Project description
The maritime industry faces ambitious international targets to reduce emissions, driven by both legislative mandates and economic incentives. Companies that improve emissions performance gain competitive advantages, such as market access and reduced costs from taxes or penalties. Consequently, there is substantial interest within the research community in developing practical solutions to demonstrably reduce emissions in maritime operations.
One prominent area of research is propulsion technology, including alternative fuels and electric propulsion systems. While these technologies may offer significant emission reductions, their adoption faces considerable barriers, such as high fleet investment costs and the need for extensive infrastructure.
However, substantial emission reductions can be achieved through low-cost, infrastructure-independent measures —namely, optimizing how ships are driven, regardless of their propulsion technology. This proposed research focuses on developing intelligent autopilot systems capable of real-time optimization of ship states —rudder position, power settings, and ballast— to align with specific route, operational, and weather conditions. This could provide very rapidly an efficient, cost-effective, and impactful solution for reducing maritime emissions.
The project will explore the application of generative models combined with reinforcement learning to enhance autopilot systems in maritime navigation. The primary goal is to optimize real-time vessel sate during navigation, reducing emissions while maintaining high levels of operational efficiency.
The research will align with and gain valuable insights from the University of Plymouth
Maritime Simulation Laboratory
's broader maritime-focused studies. The successful candidate will become an integral part of the laboratory’s research team, contributing to and benefiting from its expertise and ongoing projects.
