Real-time risk assessment of autonomous ships based on STPA and FTPN

The development of Maritime Autonomous Surface Ships (MASS) is an important trend today, and their safe navigation requires real-time risk assessment for effective risk control. The aim of this paper is to build a framework for online risk modeling of MASS to regulate the safe navigation of MASS in real-time. The framework is based on the System-Theoretic Accident Model and Process (STAMP) to build the safety control structure of MASS, using System-Theoretic Process Analysis (STPA) to identify losses or accidents and system-level hazards, identify unsafe control actions and analyze loss scenarios; consider accidents as emergence of the system, consider the results of STPA as possible system states, construct an accident model for system state transitions, and transform it into a Fuzzy-timing Petri Net (FTPN). The trapezoidal fuzzy time interval was used to represent the uncertainty time of system state transitions, the calculation method of each element in the net was given, the temporal modeling and possibility analysis method of temporal phenomena in system state transitions using the net was discussed, and a new safety level expression was introduced to measure the risk level of the system in a specific moment and situation. The final example study shows that the real-time risk assessment feedback of the method can provide a reference for the navigation supervision and risk control of MASS.