OTC-Marikultur & OTC-Mariculture2

The joint projects OTC Mariculture and OTC Mariculture2 are part of the OTC Rostock Future Cluster and address key challenges in modern aquaculture: efficiency, sustainability, and animal welfare.

Building on the results of the OTC Mariculture project, OTC Mariculture2 is developing an autonomous, AI-based system for monitoring, control, and remote management of aquaculture facilities. While the predecessor project focused on monitoring net pens in mariculture, the approach is now being extended to additional facility types and fish species. The goal is a scalable, universally applicable management system for various forms of aquaculture production.

How can aquaculture facilities be operated more efficiently, sustainably, and with greater focus on animal welfare?

OTC Mariculture2 combines sensor technology, camera systems, and artificial intelligence to enable predictive, data-driven control of aquaculture processes — from feeding to facility management, both on land and offshore.

Project Description

Objectives

The aim of OTC Mariculture2 is to develop an AI-based management system that continuously monitors aquaculture facilities, analyzes operational conditions, and actively supports operators in their decision-making.

To achieve this, sensor, image, and video data from various aquaculture systems are collected, intelligently analyzed, and translated into concrete recommendations and control actions. Special focus is placed on:

• Dynamic, resource-efficient feeding
• Increasing operational efficiency
• Ensuring animal welfare and system stability

The developed solutions are gradually tested in net pens, land-based recirculating aquaculture systems (RAS), and floating RAS systems. This creates a transferable approach adaptable to different aquaculture production environments.

Research Focus of Fraunhofer IGD

Within the project, Fraunhofer IGD focuses on AI-based analysis of image and video data for aquaculture.

At the core are computer vision and machine learning methods that enable automated, non-invasive detection of relevant animal welfare indicators and breeding parameters. These include:

• Analysis of fish behavior (e.g., feeding and schooling behavior)
• Assessment of health status and stress indicators
• AI-supported size estimation

A central research objective is the transferability of the algorithms to different fish species, facility types, and underwater as well as above-water camera systems. The developed methods are integrated as software sensors into the overarching measurement, control, and regulation system, forming the foundation of the project’s predictive management approach.