AR Ship Route Predictor Using Binocular Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality technologies are unable to accurately identify and predict the movements of maritime vehicles, such as ships, due to their unique operational capabilities and environmental influences, which poses challenges for maritime navigation safety.
Innovation Solution
A method and system that utilize binocular camera visual data to obtain feature point clouds of physical objects in a body of water, perform instance segmentation, and apply deep learning algorithms to predict the movement intentions of ships, which are then displayed on an augmented reality device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional augmented reality tracking methods are used for maritime vehicles, then the system can provide basic tracking functionality, but the accuracy of movement prediction is insufficient due to unique ship operational capabilities and environmental influences
Solution Approach 1:
The system dynamically adapts to maritime conditions by continuously adjusting tracking parameters based on environmental factors (wind, currents) and ship-specific operational characteristics. The prediction model evolves in real-time to account for the unique maneuvering patterns of maritime vehicles that differ from ground vehicles, thereby improving both prediction accuracy and tracking reliability simultaneously.
Solution Approach 2:
The system changes key prediction parameters to account for maritime-specific factors. Instead of using ground-vehicle-based maneuvering models, the system implements ship-specific parameters including environmental influence coefficients for wind and currents, and adjusted response time constants that reflect the slower, more deliberate nature of maritime vehicle operations. This parameter adaptation resolves the contradiction between prediction accuracy and reliability.
2Measurement precision
If the system processes binocular camera visual data through deep learning algorithms to predict ship movements, then prediction accuracy improves, but computational complexity and processing time increase
Solution Approach 1:
The deep learning processing pipeline is segmented into distinct modular stages: binocular visual data acquisition, feature point cloud extraction, instance segmentation, movement intention prediction, and AR display integration. Each module processes a specific aspect of the data independently, which reduces overall computational complexity while maintaining high prediction accuracy through specialized processing at each stage.
Solution Approach 2:
The system performs preliminary processing of binocular camera data by pre-extracting feature point clouds and performing instance segmentation before the main prediction algorithm executes. This preliminary action prepares the data in advance, reducing the computational burden on the prediction model and enabling faster, more efficient processing without sacrificing accuracy.
3Reliability
If the system provides real-time prediction of ship movements in confined water spaces, then navigation safety improves, but the system requires sophisticated processing that may delay response time
Solution Approach 1:
The system maintains continuous tracking and prediction of ship movements rather than performing periodic updates. The deep learning model operates continuously on the stream of binocular camera data, ensuring that prediction information is always current and available immediately when needed for navigation decisions in confined water spaces, eliminating response delays.
Solution Approach 2:
By pre-processing visual data into feature point clouds and performing instance segmentation in advance, the system prepares prediction-ready data structures continuously. This preliminary action ensures that when prediction is needed for safety-critical decisions, the computational work has already been partially completed, minimizing the time required for final prediction output.
Data Source
AI summary
According to one embodiment, a method, computer system, and computer program product for predicting the movement intentions of ships is provided. The present invention may include obtaining binocular camera visual data of a body of water; identifying physical objects in the body of water; generating feature point clouds of the physical objects in the body of water; performing instance segmentation on the generated feature point clouds; analyzing the generated feature point clouds; predicting the movement intentions of identified ships in the body of water based on the analyzed feature point clouds; and displaying the predicted movement intentions of the identified ships in the body of water on an augmented reality device.