IEEE Transactions on Instrumentation and Measurement · 2020 · 105 citations · 30 references
Artificial IntelligenceEngineeringShip ManeuveringMachine LearningMarine EngineeringIntelligent SystemsData ScienceSpace VehiclesSystems EngineeringMotion PredictionModeling And SimulationRobot LearningShip Docking—integrationShip Dynamic ModelPrediction ModellingSupervised Machine LearningPredictive AnalyticsDynamic PositioningShip ResistanceMl ModelAutomatic ControllersVessel Traffic ServiceAerospace EngineeringShip DesignSeakeeping And Control
Docking is challenging because of environmental disturbances, nearby vessels, and ship speed, and although automatic controllers are used in transit, operators still perform docking, so a dynamic model that captures vessel motion is needed. The study proposes an onboard support tool that integrates a supervised machine‑learning model with the ship dynamic model to provide position predictions. The method uses the ML model as a compensator of unmodeled dynamics, predicts 30 s ahead during docking, and is validated with historical data from the 29‑m RV Gunnerus. Including the ML model significantly improves prediction accuracy.
While automatic controllers are frequently used during transit operations and low-speed maneuvering of ships, ship operators typically perform docking maneuvers. This task is more or less challenging depending on factors, such as local environment disturbances, the number of nearby vessels, and the speed of the ship as it docks. This article proposes a tool for onboard support that offers position predictions based on an integration of a supervised machine learning (ML) model of the ship into the ship dynamic model. The ML model is applied as a compensator of the unmodeled behavior or inaccuracies from the dynamic model. The dynamic model increases the amount of predetermined knowledge about how the vessel is likely to move and, thus, reduces the black-box factor typically experienced in purely data-driven predictors. A prediction horizon of 30 s ahead of real time during docking operations is examined. History data from the 29-m coastal displacement ship Research Vessel (RV) Gunnerus are applied to validate the approach. Results show that the inclusion of the data-based ML model significantly improves the prediction accuracy.
30