Related Experiment Video
Updated: Jul 13, 2026

Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster Nephrops norvegicus
Published on: April 8, 2019
Autonomous Marine Robot Based on AI Recognition for Permanent Surveillance in Marine Protected Areas
J Carlos Molina-Molina1, Marouane Salhaoui1,2, Antonio Guerrero-González1
1Department of Automation, Electrical Engineering and Electronic Technology, Universidad Politécnica de Cartagena, Plaza del Hospital 1, 30202 Cartagena, Spain.
An autonomous surface vehicle (ASV) uses AI-powered image recognition for marine protected area (MPA) surveillance. A smart algorithm optimizes AI technology selection for real-time vessel detection and combating illegal activities at sea.
Area of Science:
- Marine conservation technology
- Artificial intelligence in environmental monitoring
- Autonomous systems for ocean surveillance
Background:
- Marine protected areas (MPAs) are vital for biodiversity but face threats from illegal activities like poaching.
- Current surveillance methods struggle with offshore communication limitations and real-time data processing.
- Existing AI solutions for vessel detection may face challenges with accuracy and latency in remote marine environments.
Purpose of the Study:
- To develop an autonomous surface vehicle (ASV) system for enhanced surveillance of marine protected areas (MPAs).
- To integrate artificial intelligence (AI)-based image recognition for detecting and identifying vessels engaged in illegal activities.
- To propose a smart algorithm for optimizing AI technology selection (cloud vs. edge) based on real-time mission requirements.
Main Methods:
- Utilized an autonomous surface vehicle (ASV) equipped with AI-based image recognition for vessel detection.
- Implemented both cloud and edge AI computing technologies for computer vision tasks, leveraging Azure services.
- Developed a smart algorithm for autonomy optimization (SAAO) to dynamically select the most suitable AI processing (cloud or edge) based on accuracy and latency needs.
Main Results:
- The AI-based system demonstrated accuracy and reliability in detecting and recognizing vessels.
- The SAAO algorithm effectively optimized surveillance by selecting appropriate AI technologies in real-time.
- The system addressed challenges of offshore communication latency by enabling intelligent AI model selection.
Conclusions:
- The proposed ASV system with AI-powered surveillance and the SAAO algorithm offers an effective solution for monitoring MPAs.
- Real-time decision-making by the SAAO enhances surveillance efficiency, reduces risks, and improves mission outcomes.
- This technology provides a robust approach to combating illegal maritime activities and protecting marine biodiversity.
More Related Videos
13:35Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring
Published on: June 13, 2025
09:19Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging
Published on: April 18, 2025