Marine Vessel Auto-Docking With Machine-Learned Control Parameters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing auto-docking systems for marine vessels are slower than human pilots and lack accuracy in controlling marine vessels during docking processes.

Innovation Solution

A computer-implemented method using a machine learning model to predict an initial guess of control parameters and a collision-free path, combined with an optimization function to minimize time and energy consumption, trained on datasets from human operators in various environments, to enhance auto-docking efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional optimization methods are used for auto-docking control, then the system can achieve accurate docking, but the docking time is too slow compared to human pilots

Engineering Contradiction:
Improvedocking accuracyVSAvoiddocking time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The machine learning model pre-calculates and provides an initial guess for the control parameters before the optimization process begins. This preliminary action based on patterns learned from human pilot data allows the optimization algorithm to start from a near-optimal point, significantly reducing the computational time required to achieve accurate docking while maintaining high precision.

Inventive Principle:
Principle #10Preliminary action

2Loss of time

If machine learning model is used to predict initial control parameters, then the docking time is reduced, but the system complexity increases

Engineering Contradiction:
Improvedocking timeVSAvoidsystem complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The machine learning model serves as an intermediary component that bridges the gap between raw sensor data and the optimization algorithm. By introducing this intermediate layer that provides informed initial guesses, the system achieves faster docking times while the added complexity is managed through the model's ability to process information efficiently and provide high-quality starting points for optimization.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If optimization function minimization is performed to reduce time and energy consumption, then the docking efficiency is improved, but the computational complexity increases

Engineering Contradiction:
Improvedocking efficiencyVSAvoidcomputational complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

By providing the optimization function with an initial guess that has already been pre-processed and refined by the machine learning model, the computational complexity of the minimization process is reduced. The optimization algorithm starts from a point close to the optimal solution, requiring fewer iterative calculations to achieve convergence, thus improving docking efficiency while managing computational demands.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4411505B1Auto-docking of marine vessels
Publication Date: 2025.10.29 ABB (SCHWEIZ) AG
  • EP4411505B1 patent drawingFigure 1
  • EP4411505B1 patent drawingFigure 2
  • EP4411505B1 patent drawingFigure 3

AI summary

The present invention generally relates to a computer-implemented method for generating control parameters for auto-docking of a marine vessel (10), and to a control unit (16) for executing the method, to a marine vessel comprising the control unit, and to a corresponding computer program product.