Incorporating Approximate Dynamics Into Data-Driven Calibrator: A Representative Model for Ship Maneuvering Prediction

Abstract

High-fidelity models capable of accurately predicting ship motion are critical for promoting innovation and efficiency in the maritime industry. However, creating an advanced model that comprehensively represents the system and its interaction with dynamic environments has always been challenging. Many models provide partial knowledge about a system. To handle the deficiency and improve model fidelity, we propose a hybrid modeling methodology, in which prior knowledge describing the ship dynamic effects is incorporated into a data-driven calibrator, yielding a representative model with high predictive capability. Enabled by the integration of model estimated ship states into the calibrator, the informative information could be interpreted and carried forward. Simulation and full-scale experiments are conducted on the research vessel Gunnerus to exemplify the concept. A best available numerical model and a neural network are prepared to be the foundation and calibrator, respectively. Experiment results show that the cooperative model greatly improves the predictive capability of the research vessel. From the ship modeling perspective, this study provides new insights by bridging the gap between two separate domains: model-based and data-driven.

A Review on Motion Prediction for Intelligent Ship Navigation

“…Alongside the accumulation of ship maneuvering control and navigation data, apart from the model ( 1)-(3) mentioned earlier, black-box models based on neural networks, data, and similar techniques are gradually finding applications in ship motion prediction and navigation control [77][78][79][80]. Black-box models can take into account a wider range of inputs, leading to higher predictive accuracy.…”

Evidence from China's shipping sector on the impact of fiscal measures in enabling a low-carbon economic transition

“…The Chinese maritime industry's response to a one-standard-deviation shift in TBPs is manifested through the LCT [ 30 , 31 ]. The impact gradually becomes more consistent over ten periods, specifically after the initial five and the concluding five, which exhibit a linear pattern.…”

Knowledge and data in cooperative modeling: Case studies on ship trajectory prediction

“…On the other hand, Skulstad et al (2021a) trained a neural network with onboard sensor measurements so that it compensates for the multiple-step-ahead position error made by the physics-based model. In Wang et al (2021), Wang et al proposed a data-driven model that maps the future position calculated by the similar ship's dynamic model into that of the targeting ship. In their approaches, the roles of the physics-based and data-driven models are clearly distinguished, thus contributing to better interpretability and maintainability of the cooperative architecture.…”

“…In addition, in the full-scale experiment, having too much uncertainty from unexpectable environmental disturbances in the longer prediction horizon makes a fair comparison between dynamic models challenging. Due to this limitation, errors over the prediction horizon from 15 s to 30 s have been widely used as a metrics of the accuracy of ship dynamic models in the full-scale experiment (see Skulstad et al (2021a) and Wang et al (2021)). The zigzag maneuver is one of the maneuvers that the International Towing Tank Conference (the ITTC) recommends as a full-scale maneuvering trials procedure.…”

“…Recently, the maritime applications are not exceptions in a trend of making a synergy of scientific knowledge and data (e.g., Fonseca and Gaspar (2021)). Skulstad et al (2021a) and Wang et al (2021) presented breakthrough ideas that compensate/map trajectories made by the available physics-based model into true trajectories in a datadriven manner. Such data-driven geometrical compensation/mapping achieved a good performance while keeping the stability of the physicsbased model untouched as a stable foundation of the model.…”