BEHAVIOR MODELING AND EMERGENCY SIMULATION FOR A MARITIME AUTONOMOUS VESSEL WITH TRAJECTORY STABILITY ASSESSMENT

https://doi.org/10.33815/2313-4763.2025.1.30.195-208

Abstract

The article investigates the complex behavior of a Maritime Autonomous Surface Ship (MASS) during critical emergency situations. It specifically focuses on scenarios involving loss of control, sudden gusty winds, and steering drive failure. It is demonstrated that studying the emergency behavior of a MASS holds profound practical significance, directly addressing challenges related to the safe deployment and energy efficiency of autonomous navigational transits.

A simplified mathematical model, based on Newton-Euler equations, was employed to characterize the vessel's motion dynamics in the horizontal plane. This model accounts for hydrodynamic, aerodynamic, and stochastic effects in a simplified manner, thereby ensuring a realistic representation of the marine environment.

A critical gap identified in current MASS research is the deficit of models capable not only of reproducing the behavior of autonomous vessels during rudder blockage, loss of control signals, or unexpected disturbances, but also of predicting their cascading consequences. To address this deficiency, an innovative simulation scenario for MASS emergency modes was proposed. This scenario incorporates various navigational element failures. The simulation was conducted in MATLAB/Simulink, enabling the implementation and analysis of key emergency scenarios. This included a discrete integration step of 0.1 seconds over a 100-second simulation period, utilizing typical parameters for a small autonomous platform.

The study evaluated three distinct emergency scenarios:

– in the first scenario, the rudder was fixed at a zero position, leading to irreversible and uncontrolled course deviation due to wind gusts. This was identified as the most critical outcome, as the vessel loses its ability to perform navigational compensation, although the direction of drift remains predictable;

– the second scenario involved simulating a delay in the PID controller's command signal, mimicking an unstable connection. This revealed deviations from the desired course, indicating a partial retention of trajectory. However, it underscored that prolonged delays can induce not only aperiodic effects but also resonance, thereby necessitating adaptive control algorithms;

– in the third scenario, with the PID controller actively engaged, the system demonstrated a robust capability to quickly compensate for a strong lateral wind gust. The vessel's course stabilized within a few seconds after a minor transient overshoot, confirming a partial invariance to disturbances.

The simulation results provided invaluable insights into trajectory stability and the varying effectiveness of control strategies under adverse conditions. The presented approach to simulating critical states, achieved by fixing motion parameters during loss of control, established a fundamental database for developing proactive crisis response protocols. Furthermore, the classic PID control model was enhanced with a "freeze" function for actuators and adapted to account for changes in load and transverse speed. This enables the creation of redundant autonomous control loops, potentially through digital twin technologies. These data serve as a foundation for developing new automatic emergency response algorithms aimed at mitigating unforeseen and critical situations during future autonomous navigation missions.