A single instrumented fish tank, packed with cameras, weight sensors, and water-quality probes, will test whether a modular, land-based aquaculture system can monitor fish welfare in real time. Most fish farming today relies on visual checks and water sampling, which miss early signs of stress or disease. This project aims to close that gap by embedding multiple sensors directly into a compact recirculating system—the Micro Modular Recirculating Aquaculture System (MM-RAS). The sensors track fish behaviour, growth, and water conditions continuously, while AI models analyse the data to flag problems before they become visible to a human operator. If the feasibility study succeeds, the system could give fish farmers a practical, automated tool to improve welfare without increasing labour. The cloud-based dashboard would let operators monitor fish counts, water quality, and energy use from anywhere. For the UK Food Valley initiative, this technology supports sustainable, high-welfare aquaculture that uses less water and land than traditional methods. The project also produces an investor pitch deck and commercialisation strategy, preparing the system for market entry if the pilot validates its performance and scalability.
View original technical description
This project advances the Micro Modular Recirculating Aquaculture System (MM-RAS) to bring an efficient, scalable, and welfare-oriented aquaculture solution closer to commercialisation. Building on prior successes, we aim to refine and validate the MM-RAS system through incremental improvements in system design, integrated monitoring, and commercial readiness. Our work supports the UK Food Valley initiative by developing technology that addresses sustainable food production, resource efficiency, and high standards of fish welfare. Through six work packages, we will implement and test a single instrumented breeding system equipped with integrated cameras, weight sensors, in-water sensors, and non-contact sensors to generate key data on fish health, water quality, and system performance. In WP1, we will refine hardware and software design to ensure compatibility and scalability. WP2 will collect, validate, and analyse welfare and performance data, creating datasets essential for system calibration and future enhancements. WP3 develops AI and machine learning models to monitor fish health and operational metrics in real time, WP4 develops data science from wet and dry sensors using a cloud-based dashboard to make insights accessible. WP5 then validates the system's performance and welfare impact in trials, providing recommendations for optimising the design. WP6 coordinates these efforts, delivering regular reports, an investor pitch deck, and a commercialisation strategy. Our goal is to establish the MM-RAS as a high-performing, welfare-focused aquaculture system with commercial viability. We wish to demonstrate the feasibility of a modular platform that supports both hardware and software integration to track and improve fish welfare, optimise resource usage, and enable automation. The project will provide a foundation for future investor engagement, with clear data on system value and scalability. **Aim:** To advance the Micro-Modular Recirculating Aquaculture System (MM-RAS) through integrated monitoring, data analytics, and commercialisation strategies, to demonstrate system viability, scalability, and feasibility. **Objectives:** 1. Integrate monitoring technologies (cameras, sensors) into a test case to assess fish health, welfare, and system performance. 2. Collect data on water usage, energy efficiency, and fish growth to build a value proposition for investor pitches and external funding. 3. Develop and deploy a scalable, cloud-based dashboard for real-time monitoring of fish counts, water quality, and operational efficiency. 4. Conduct a pilot study to validate system readiness, focusing on automation, animal welfare, and system scalability. 5. Explore potential revenue models and investor readiness through the development of an investor pitch deck and third-party validation.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know