If you are a food distribution company dealing with high staff turnover, tight delivery windows, and the challenge of automating warehouses where people still need to walk — this project developed self-deploying robot fleets that pick, palletize, and unwrap goods in real food warehouses. The final system was demonstrated at the National Centre for Food Manufacturing with all functionalities working in a live environment alongside human workers.
Self-Deploying Robot Fleets for Warehouse Automation That Work Safely Alongside People
Imagine a warehouse where a team of robots can show up, learn the layout on their own, and start moving goods — all while dodging forklifts and workers without anyone programming every route. That's what ILIAD built: robot fleets that deploy themselves in real warehouses, pick and stack products, and get smarter over time by watching how people move around them. They tested everything in a real food distribution centre, not just a lab. The robots coordinate with each other like a well-drilled crew, splitting tasks and replanning on the fly when things change.
What needed solving
Warehouses — especially in food distribution — face a brutal combination: labour shortages, rising wages, faster delivery expectations, and the need to keep human workers safe around moving equipment. Traditional automation (conveyor belts, fixed AGVs) requires months of installation and can't adapt when layouts change or demand spikes. Companies need robots that can show up, learn the space, work safely around people, and coordinate as a team — without reprogramming everything each time.
What was built
The project delivered self-deploying robot fleets that can enter a new warehouse, map it autonomously, and begin picking, palletizing, and unwrapping goods — all while safely operating around human workers. The final system was demonstrated at the National Centre for Food Manufacturing with all functionalities working in a real environment, including fleet management, human-awareness, and long-term self-optimisation across 20 total deliverables.
Who needs this
Who can put this to work
If you are a 3PL provider that constantly reconfigures warehouse operations for different clients — this project developed robots that self-deploy into new environments without manual mapping or programming. The system handles heterogeneous robot fleets with integrated task allocation and motion planning, meaning you can drop robots into a new facility and they figure out the layout and workflows themselves.
If you are an e-commerce fulfilment operator struggling to scale picking operations during demand peaks while keeping workers safe — this project developed fleet management with formal safety guarantees for mixed human-robot environments. The system uses deep learning to track and predict human movement patterns, enabling robots to operate efficiently without creating safety hazards in busy warehouse aisles.
Quick answers
What would it cost to deploy this robot fleet system in our warehouse?
The project data does not include specific pricing or per-unit costs. The system was designed to integrate with existing warehouse facilities rather than requiring a full rebuild, which should reduce deployment costs compared to fixed automation. Contact the consortium partners for commercial licensing terms.
Can this scale to a full-size warehouse with hundreds of robots?
The system was designed for scalable fleets of heterogeneous robots — meaning different robot types working together. The final demonstration at the National Centre for Food Manufacturing validated all core functionalities including fleet management with formal guarantees, though the exact number of simultaneous robots in testing is not specified in the available data.
Who owns the IP, and can we license this technology?
The consortium includes 10 partners across 4 countries, with 5 industry partners and 3 SMEs alongside 5 universities. IP is likely split across partners per EU funding rules. Örebro University coordinated the project. Contact individual partners for licensing specific components like the self-deployment system, fleet management, or human-awareness modules.
Has this been tested in a real warehouse, not just a lab?
Yes. The project ran three major demonstrations, progressing from simulation at Örebro University to a real food distribution environment at the National Centre for Food Manufacturing (NCFM). The final demonstration included self-deployment, long-term operation, human-awareness, fleet management, picking, palletizing, and unwrapping — all in a real facility.
How does this handle safety around human workers?
ILIAD specifically tackled human safety in mixed environments, providing foundations for future safety standards. The system tracks and analyses human movement, predicts future activity patterns from long-term observations, and plans socially normative robot movements using learned human behaviour models. This goes beyond simple collision avoidance.
How long does it take to deploy robots in a new warehouse?
The key selling point is self-deployment — robots learn the environment rather than requiring manual programming of every route and location. Based on the project objectives, the system uses long-term observation to build and improve its environmental models over time, meaning it gets more efficient the longer it operates. Specific deployment timelines are not stated in the available data.
Does this work with our existing warehouse management system?
The project was designed to integrate with current warehouse facilities rather than replace them. The system includes online integrated task allocation, motion planning, and coordination for heterogeneous robot fleets, suggesting it can work alongside existing infrastructure. Specific WMS integrations would need to be discussed with the technology partners.
Who built it
The ILIAD consortium is well-balanced for bringing warehouse robotics to market: 10 partners split evenly between 5 industry players and 5 universities across 4 countries (Germany, Italy, Sweden, UK). The 50% industry ratio is strong for a research project, and with 3 SMEs in the mix, there are agile companies positioned to commercialize components. Örebro University coordinated — they are a leading European robotics research centre. The inclusion of the National Centre for Food Manufacturing at the University of Lincoln gave the project access to a realistic test site, which is why the final demonstration could happen in a real food distribution environment rather than a lab. For a business looking to adopt this technology, the industry partners are the most likely route to commercial products or licensing deals.
- OREBRO UNIVERSITYCoordinator · SE
- UNIVERSITY OF LINCOLNparticipant · UK
- ROBERT BOSCH GMBHparticipant · DE
- TECHNISCHE UNIVERSITAET MUENCHENparticipant · DE
- UNIVERSITA DI PISAparticipant · IT
- GOTTFRIED WILHELM LEIBNIZ UNIVERSITAET HANNOVERparticipant · DE
Örebro University (Sweden) coordinated this project. Their robotics department (AASS Research Centre) is publicly listed. Industry partners may offer commercial licensing.
Talk to the team behind this work.
Want an introduction to the ILIAD team or help identifying which partner has the component you need? SciTransfer can connect you directly — contact us for a matchmaking consultation.