Jove
Visualize
Contact Us

Related Concept Videos

Real-World Applications of Space Curves01:29

Real-World Applications of Space Curves

Modern aerospace navigation depends on the accurate prediction of motion in three-dimensional space. In defense applications, radar systems continuously track both interceptors and moving aerial targets to find whether their flight paths will result in a collision. These motions are modeled mathematically as space curves, which represent paths that change continuously with time. Each object’s position is described by a vector function that specifies its location in terms of time-dependent...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Advancements in prostate cancer segmentation: Integrating prostate zonal information.

Digital health·2026
See all related articles
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 6, 2026

Trajectory Data Analyses for Pedestrian Space-time Activity Study
16:14

Trajectory Data Analyses for Pedestrian Space-time Activity Study

Published on: February 25, 2013

13.5K

Enhancing Maritime Safety: Estimating Collision Probabilities with Trajectory Prediction Boundaries Using Deep

Robertas Jurkus1, Julius Venskus2, Jurgita Markevičiūtė2

  • 1Institute of Data Science and Digital Technologies, Vilnius University, 01513 Vilnius, Lithuania.

Sensors (Basel, Switzerland)
|March 17, 2025
PubMed
Summary

This study uses deep learning to forecast vessel paths and assess collision risks near Bornholm Island. Conformal Prediction Regions (CPR) reliably predict maritime collision risks, enhancing navigational safety.

Keywords:
collision risk scoreconformal prediction regionslong short-term memoryuncertainty quantificationvessel collision detectionvessel trajectory prediction boundaries

More Related Videos

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.3K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.6K

Related Experiment Videos

Last Updated: Jul 6, 2026

Trajectory Data Analyses for Pedestrian Space-time Activity Study
16:14

Trajectory Data Analyses for Pedestrian Space-time Activity Study

Published on: February 25, 2013

13.5K
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.3K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.6K

Area of Science:

  • Maritime Safety
  • Artificial Intelligence
  • Deep Learning

Background:

  • Maritime accidents pose significant risks, necessitating advanced safety strategies.
  • Accurate vessel trajectory forecasting and collision risk assessment are crucial for prevention.
  • Existing methods may not fully capture the complexities of real-world maritime scenarios.

Purpose of the Study:

  • To develop and evaluate an AI/ML-based approach for forecasting vessel trajectories.
  • To assess maritime collision risks using integrated deep learning and statistical methods.
  • To improve navigational decision-making and maritime risk management protocols.

Main Methods:

  • Utilized Long Short-Term Memory (LSTM) autoencoders, a type of deep recurrent neural network, for trajectory prediction.
  • Integrated trajectory forecasts with statistical techniques to create probabilistic boundaries (confidence intervals, prediction intervals, ellipsoidal prediction regions, conformal prediction regions).
  • Developed a collision risk score based on the likelihood of probabilistic boundary overlaps.

Main Results:

  • Applied the methodology to simulated scenarios and the 2021 Scot Carrier-Karin Hoej collision case study.
  • Demonstrated that Conformal Prediction Regions (CPR), a non-parametric method, reliably forecast collision risks with 95% confidence.
  • Showcased the effectiveness of integrating statistical uncertainty quantification with deep learning models.

Conclusions:

  • The proposed AI/ML approach significantly enhances the accuracy of collision risk assessment.
  • Conformal Prediction Regions offer a robust, non-parametric solution for reliable maritime risk forecasting.
  • Findings support a shift towards proactive, AI/ML-enhanced maritime risk management for improved safety.