Ship Pitch Prediction Using Camera-Based Throttle Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ship assistance technologies do not effectively predict and mitigate pitching in small ships, which can lead to discomfort for occupants and require manual throttle adjustments.
Innovation Solution
A ship assistance device that calculates the pitching amount of a ship body based on images from a camera, estimates the pitching cycle, predicts the pitching behavior, and automatically controls the throttle to reduce pitching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If pitching control is performed based on angular velocity sensor output exceeding a threshold, then pitching response is achieved, but pitching cannot be predicted in advance
Solution Approach 1:
The system performs preliminary action by predicting pitching motion in advance using image processing and cyclic pattern recognition. The pitching amount calculation unit calculates pitching amount from sequential images, the pitching cycle estimation unit estimates the cyclic pattern, and the pitching prediction unit predicts future pitching based on the estimated cycle, enabling proactive throttle control before pitching occurs
Solution Approach 2:
The system implements feedback by continuously monitoring actual pitching from image processing, comparing it with predicted pitching, and adjusting throttle control accordingly. The pitching amount calculation unit provides continuous feedback on actual ship body attitude, which is used to refine the pitching cycle estimation and improve prediction accuracy over time
2Measurement precision
If image processing is performed to calculate pitching amount, then measurement precision is improved, but processing load increases
Solution Approach 1:
The system extracts only the essential information needed for pitching measurement from images. The pitching amount calculation unit identifies specific feature points or regions in sequential images that directly correlate with ship body attitude, extracting only the necessary visual data rather than processing entire images, thereby reducing computational load while maintaining measurement precision
Solution Approach 2:
The system uses visual copying of ship body orientation from images as a substitute for direct sensor measurement. By capturing and analyzing sequential images of the ship body or horizon reference, the system creates a visual copy of the pitching state that can be processed more efficiently than using complex angular velocity integration
Data Source
AI summary
A ship assistance device including a storage medium storing a computer-readable command and a processor connected to the storage medium, the processor executing the computer-readable command to: calculate a pitching amount of a ship body based on a plurality of images photographed by a camera mounted on the ship body; estimate a pitching cycle of the ship body at least based on the calculated pitching amount; predict pitching of the ship body based on the estimated pitching cycle; and control a throttle of the ship body so as to reduce the predicted pitching of the ship body.