About the project
How should an autonomous ship respond when navigational risk is uncertain and continuously changing? This project will develop risk-aware decision-making methods that combine predicted traffic situations, navigational risk, uncertainty, international regulations for preventing collisions at sea (COLREGs) and operational constraints to support safe navigation decisions, including collision avoidance, operational adjustment, supervision and fallback.
Autonomous ships must make navigation decisions in complex environments where the behaviour of other vessels, future traffic developments and associated risks are uncertain. Avoiding collision alone is not sufficient: decisions must also consider navigational rules, operational constraints, safety margins and the consequences of uncertainty.
This project will investigate risk-aware decision-making for safe autonomous navigation. The research will explore how dynamic navigational risk can be incorporated explicitly into autonomous decision processes, enabling navigation actions to respond to both the estimated level of risk and confidence in that assessment.
The project will investigate decision-making during vessel encounters, including crossing, head-on and overtaking situations. Potential actions may include maintaining course and speed, altering course, reducing speed or increasing separation. The research will examine how these decisions can balance navigational safety, COLREGs compliance and operational efficiency, particularly when the intentions or future trajectories of surrounding vessels remain uncertain.
The project will also consider situations in which continued autonomous operation may no longer be appropriate. Methods will be investigated for linking changing risk conditions to operational responses such as increased safety margins, function degradation, remote supervision, mode switching or fallback. Potential approaches include:
- probabilistic decision-making
- multi-objective optimisation
- risk-constrained planning
- model predictive methods
- machine learning
Methods will be evaluated using representative maritime traffic scenarios and simulation environments.
The project will advance autonomous navigation from collision avoidance towards risk-informed decision-making, contributing to safer and more explainable autonomous ship operations under complex and uncertain conditions.
The School of Engineering is committed to promoting equality, diversity inclusivity as demonstrated by our Athena SWAN award. We welcome all applicants regardless of their gender, ethnicity, disability, sexual orientation or age, and will give full consideration to applicants seeking flexible working patterns and those who have taken a career break. The University has a generous maternity policy, onsite childcare facilities, and offers a range of benefits to help ensure employees’ well-being and work-life balance. The University of Southampton is committed to sustainability and has been awarded the Platinum EcoAward.
