Expected Outcome:
Project results are expected to contribute to the following expected outcomes:
- Development of closed-loop scientific experimentation systems that integrate automation with AI-driven, trustworthy decision-making processes in existing laboratory environments;
- Accelerated scientific discovery with increased efficiency and reproducibility;
- Improved scientific productivity;
- Advancement of laboratory automation, including development of best practices, challenges, and opportunities for accelerating R&D;
- Prototype functional demonstrators that showcase the integration of automation with AI-driven decision-making, enabling the development of closed-loop scientific experimentation systems.
Scope:
This topic addresses the development of safe and trustworthy closed loop scientific experimentation systems through the integration of laboratory automation with AI-driven decision-making processes and robust data infrastructures. The funded project will help scientific labs with an already advanced level of automation and digitalisation to design, develop, and test the intelligence layer that enables scientific instrumentation to semi- or fully autonomously plan, run, and analyse experiments, ideally in coordination/network with other labs and without requiring a complete redesign of existing laboratory outfitting.
Proposals will incorporate comprehensive data management systems capable of handling the collection, storage, processing, and sharing of experimental data. This includes developing scalable and secure data storage solutions, efficient data processing and analysis tools, and mechanisms to facilitate data sharing and collaboration across labs, while ensuring data security and privacy.
Systems could incorporate AI-driven resource optimisation modules, actively minimising energy, reagent, and material consumption during automated experimentation cycles. Systems should incorporate appropriate level of security and robustness by design.
Proposals should demonstrate how an existing lab can be retrofitted with AI-driven software systems to plan, execute, and analyse experiments in a closed-loop fashion, incorporating human oversight and interaction to ensure accuracy, safety, and ethical compliance.
Possible research targets include (non-exhaustively):
- Autonomous/semi-autonomous and adaptive AI systems (including agentic AI) that connect with laboratory instruments and robotics and can autonomously plan, act, learn and adapt within a scientific environment, within a validated safe pipeline;
- Assistive and interactive safe AI-managed robotic systems that automate diverse experiments and can be applied to a diverse hardware setup;
- Scalable automation solutions and networked AI systems that enable collaborative experimentation across multiple labs and networks of labs (including different geographic locations), supporting the simultaneous execution of large volumes of experiments;
- Systems that provide real-time data processing and analytics, enabling immediate feedback and dynamic adjustments during experiments;
- Standards and protocols to ensure interoperability between different laboratory instruments, robotics, and AI systems;
- Intuitive user interfaces for enhanced human-machine interaction;
- AI-driven predictive maintenance systems to optimize equipment uptime and resource utilization;
- Exploration of how scientific automation technologies can be adapted for use in various scientific disciplines beyond those in the scope of this call.
While the scope of this call prioritises software development, it does not exclude the justified purchase of complementary equipment necessary to implement the research targets of the project.
Impact areas of automated experimentation in biomass and precision fermentation could include the development of alternative protein sources in food production and alternative fats, bio-based materials, specialty carbohydrates, biotechnologies in food systems (such as biochemicals, microbial cultures, etc), food ingredients.
International collaboration is encouraged.
