An.Dy · Project

Smart Robots That Predict Worker Movements for Safer, Faster Factory Collaboration

manufacturingTestedTRL 5

Imagine working alongside a robot that can guess your next move — like a dance partner who never steps on your toes. The An.Dy project built a wearable suit that tracks how your whole body moves, a brain for the robot that learns your habits, and control software that lets the robot physically help you before you even ask. They tested this with factory cobots, wearable exoskeletons that reduce back strain, and even humanoid robots that can hand you heavy parts while keeping both of you balanced.

By the numbers

EUR 3,950,025

EU research investment

consortium partners across 6 countries

validation scenarios tested (cobot, exoskeleton, humanoid)

44%

industry ratio in consortium

total project deliverables produced

SME partners involved

The business problem

What needed solving

Factory workers get injured when robots cannot predict their movements during physical collaboration tasks — leading to either unsafe close contact or inefficient safety distances that kill productivity. Current cobots react to accidental bumps but cannot anticipate intentional collaboration, forcing workers into awkward postures and repetitive strain. Companies need robots that work with people the way a good colleague would: reading intentions and sharing the physical load proactively.

The solution

What was built

The project built three core technologies: ANDYSUIT, a wearable suit that tracks whole-body forces and motion; ANDYMODEL, software that learns how individual workers behave during collaborative tasks; and ANDYCONTROL, a predictive control system that lets robots physically assist workers before being asked. These were validated across 3 scenarios with cobots, exoskeletons, and humanoid robots, producing 25 deliverables including 4 detailed demonstration reports.

Audience

Who needs this

Automotive assembly plants deploying collaborative robots on production linesLogistics and warehousing companies evaluating exoskeletons for manual handling workersAerospace MRO facilities where technicians co-manipulate heavy componentsCobot manufacturers looking to add predictive collaboration features to their product lineExoskeleton startups needing worker behavior modeling to improve comfort and adoption

Business applications

Who can put this to work

Automotive & Heavy Manufacturing

enterprise

Target: Automotive assembly plants and heavy manufacturers with manual lifting and collaborative robot stations

If you are an automotive manufacturer dealing with worker injuries from repetitive lifting and awkward postures on the assembly line — this project developed the ANDYSUIT wearable tracking system and ANDYCONTROL predictive software that lets cobots adapt in real time to each worker's ergonomic needs. The technology was validated in a realistic human-cobot physical collaboration scenario. With 4 industry partners already involved, the path from lab to your shop floor is shorter than starting from scratch.

Logistics & Warehousing

mid-size

Target: Warehouse operators and logistics companies investing in exoskeletons for manual handling workers

If you are a logistics company struggling with high injury rates and staff turnover from physical strain in picking and packing — this project built the ANDYMODEL, a learning system that optimizes exoskeleton assistance to each worker's comfort and reduces physical stress. The second validation scenario specifically tested robot-as-exoskeleton for human comfort optimization. The EUR 3,950,025 research investment produced wearable force-tracking technology that could cut your workers' compensation costs.

Aerospace & Precision Assembly

enterprise

Target: Aerospace MROs and precision manufacturers needing human-robot co-manipulation of heavy or delicate parts

If you are an aerospace maintenance company where technicians must hold heavy components in awkward positions during inspection or assembly — this project created predictive control that lets a humanoid robot share the physical load while maintaining balance for both human and machine. Validated across 3 distinct robot types (cobot, exoskeleton, humanoid) over a 4-year research program with 9 consortium partners, this is not a single-use trick but a transferable collaboration platform.

Frequently asked

Quick answers

What would it cost to implement this technology in our facility?

The project itself received EUR 3,950,025 in EU funding across 9 partners over 4+ years, indicating significant R&D investment. Licensing or implementation costs would need to be negotiated directly with the consortium lead, Fondazione Istituto Italiano di Tecnologia. Based on available project data, no commercial pricing has been published.

Can this scale to a full production line with dozens of workers?

The project validated its technology in realistic but controlled scenarios — human-cobot collaboration and human-humanoid collaboration. Scaling to full production lines with multiple simultaneous workers would require further engineering. The ANDYSUIT wearable system and ANDYMODEL learning component were designed for individual worker tracking, so per-worker deployment is the current baseline.

Who owns the intellectual property, and can we license it?

The consortium is led by Fondazione Istituto Italiano di Tecnologia (Italy), a non-profit research institute. IP from EU-funded RIA projects is typically owned by the partners who generated it. With 3 SMEs and 4 industry partners in the 9-member consortium, licensing discussions should start with IIT or the relevant technology developer within the consortium.

Does this comply with EU machinery and workplace safety regulations?

The project explicitly builds on Europe's leadership in safety-certified collaborative robots and extends it to intentional physical interaction. However, the deliverables describe scientific papers and scenario reports, not certification documents. Any deployment would still need to pass standard CE marking and machinery directive compliance for your specific use case.

How long would integration with our existing cobots take?

The first validation scenario specifically tested human-cobot physical collaboration with industrial collaborative robots. Based on available project data, the ANDYCONTROL system was demonstrated as a research prototype, not a plug-and-play product. Integration timelines would depend on your specific cobot platform and would likely require a pilot engagement with the research team.

What concrete results came out of the project?

The project produced 25 deliverables including 4 demonstration scenario reports covering human-cobot and human-humanoid physical collaboration. Three key technologies were built: ANDYSUIT (wearable force and motion tracking), ANDYMODEL (cognitive behavior learning), and ANDYCONTROL (predictive collaboration control). These were tested across 3 validation scenarios with different robot types.

Consortium

Who built it

The An.Dy consortium of 9 partners across 6 countries (Germany, Denmark, France, Italy, Netherlands, Slovenia) shows a balanced mix for technology transfer: 4 industry partners and 4 research organizations, with a 44% industry ratio and 3 SMEs on board. This means the research was not done in an ivory tower — nearly half the consortium had commercial incentives to make the technology practical. The project was led by Fondazione Istituto Italiano di Tecnologia, one of Europe's top robotics institutes, which adds credibility but also means the primary contact is a research body, not a commercial vendor. For a business looking to adopt this technology, the SME partners may be the fastest route to a deployable product.

FONDAZIONE ISTITUTO ITALIANO DI TECNOLOGIACoordinator · IT
INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET AUTOMATIQUEparticipant · FR
MOVELLA TECHNOLOGIES BVparticipant · NL
INSTITUT JOZEF STEFANparticipant · SI
ISTITUTO NAZIONALE ASSICURAZIONE INFORTUNI SUL LAVORO INAILthirdparty · IT
IMK AUTOMOTIVE GMBHparticipant · DE
DEUTSCHES ZENTRUM FUR LUFT - UND RAUMFAHRT EVparticipant · DE
ANYBODY TECHNOLOGY ASparticipant · DK
OTTOBOCK SE & CO. KGAAparticipant · DE

How to reach the team

Fondazione Istituto Italiano di Tecnologia (IIT), Italy — a leading European robotics research institute. SciTransfer can help identify the right person on the An.Dy team for your specific use case.

Order a report on this consortium See FONDAZIONE ISTITUTO ITALIANO DI TECNOLOGIA profile →

Next steps

Talk to the team behind this work.

Want to explore how An.Dy's predictive robot collaboration technology could reduce injury rates or boost throughput in your facility? SciTransfer connects you directly with the research team — contact us for a tailored briefing.

Order a report on this topic More in Manufacturing & Industry 4.0