Vessel Trajectory Adaptation Using Surrogate Wave Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vessel routing systems are inefficient due to reliance on slow and expensive weather forecasts, which can lead to increased costs, carbon footprints, and network disruptions, as they fail to provide real-time, accurate sea state information, especially in areas with communication breakdowns.
Innovation Solution
A cognitive system integrating high-performance computing for large-scale wave forecasting with lightweight surrogate models, enabling dynamic vessel trajectory planning and adaptation based on forecasted wave conditions and user-defined constraints, using a combination of machine learning and IoT data for optimized ship routing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional weather forecast systems are used for vessel routing, then comprehensive weather data can be obtained, but the system becomes slow and expensive with increased carbon footprints
Solution Approach 1:
The system segments the wave forecasting task into two parts: a comprehensive high-resolution forecast model that runs periodically to generate detailed weather data, and a lightweight surrogate model that runs locally on the vessel to provide rapid real-time predictions. This segmentation allows the system to maintain high accuracy while reducing computation time for operational decision-making.
Solution Approach 2:
The surrogate model acts as an intermediary between the comprehensive forecast system and the vessel routing decision system. It translates the complex high-resolution forecast data into rapid, actionable predictions that can be processed in real-time on the vessel, bridging the gap between detailed forecasting and operational needs.
2Measurement precision
If high-resolution wave forecasting models are used, then accurate sea state information is obtained, but computational cost and expense increase
Solution Approach 1:
The computational workload is segmented between a high-resolution forecast model that runs on ground-based supercomputing resources and a lightweight surrogate model that runs on the vessel's onboard computer. This division allows accurate sea state information to be generated without burdening the vessel's energy resources.
Solution Approach 2:
The system creates a simplified copy (surrogate model) of the comprehensive forecast model that captures the essential prediction capabilities while requiring minimal computational resources. This copy can be executed rapidly on the vessel using a fraction of the energy required by the full high-resolution model.
3Speed
If real-time wave forecasting is implemented, then rapid trajectory adaptation is enabled, but network connectivity requirements increase
Solution Approach 1:
The surrogate model serves as an intermediary that enables real-time trajectory adaptation without continuous network connectivity. It processes local sensor data and previously received forecast information to generate rapid predictions, allowing the vessel to adapt its trajectory in real-time even in areas with limited or no network access.
Solution Approach 2:
The system implements self-service capabilities by enabling the vessel to generate its own wave forecasts using the lightweight surrogate model and onboard sensors. This reduces dependence on continuous network connectivity and external forecasting services, allowing the vessel to autonomously adapt its trajectory in real-time.
Data Source
AI summary
Embodiments for implementing intelligent vessel trajectory planning and adapting by a processor. A trajectory of a vessel may be dynamically determined according to forecasted wave conditions using a surrogate wave model, a wave forecasting model, one or more user defined constraints, or a combination thereof.