SoMa · Project

Soft Robot Hands That Grip Irregular Objects Without Complex Programming

manufacturingTestedTRL 5

Imagine trying to pick up a ripe tomato with a metal claw — you'd crush it. That's basically the problem robots have today: their rigid grippers can't handle soft, oddly shaped, or delicate items without expensive sensors and complicated programming. SoMa built flexible rubber-like robot hands that naturally adapt their grip to whatever they touch, the same way your hand wraps around a coffee cup without you thinking about it. The trick is using the object itself and nearby surfaces to guide the grip, instead of trying to calculate every finger position in advance.

By the numbers

consortium partners building soft manipulation technology

countries in the research consortium (AT, CH, DE, IT, UK)

hand prototypes delivered by each of two partners (UNIPI and TUB) for industrial integration

total project deliverables produced

industrial partners in the consortium (29% industry ratio)

The business problem

What needed solving

Robotic grippers today are rigid, fragile, and require extensive programming for each object type — making automation of varied or delicate products prohibitively complex and expensive. Industries handling irregular items like fresh produce, mixed warehouse goods, or fragile components either cannot automate or face high reject rates and constant re-engineering. The gap between what robots can grip and what human hands effortlessly handle is a bottleneck for automation ROI.

The solution

What was built

The project built multiple soft robotic hand and gripper prototypes across 4 partner labs (UNIPI, DLR, TUB, IIT), enhanced with integrated sensors. Specifically, 4 prototype units each from UNIPI and TUB were delivered to OCADO for warehouse picking integration and to Disney for entertainment applications, with 23 deliverables produced in total.

Audience

Who needs this

Automated warehouse and fulfillment center operators picking mixed productsFresh produce packhouses and growers needing automated harvesting or sortingElectronics manufacturers assembling fragile or variable-shape componentsFood processing companies handling soft or irregular items on production linesRobotics integrators looking for next-generation gripper technology to offer clients

Business applications

Who can put this to work

Warehouse & Grocery Logistics

enterprise

Target: Automated fulfillment center operators

If you are a warehouse operator dealing with the challenge of automating pick-and-pack for thousands of different product shapes — this project developed soft robotic hand prototypes that were delivered directly to OCADO for warehouse integration testing. The hands adapt their grip to each object shape without needing per-item programming, which could drastically cut the engineering time needed to automate new product lines.

Fresh Produce & Agriculture

any

Target: Fruit and vegetable growers or packhouses

If you are a fresh produce company struggling with labor shortages at harvest or in sorting lines — this project built compliant grippers specifically relevant to fruit and vegetable handling, as indicated by its EuroSciVoc classification. The soft hands avoid bruising delicate produce because they naturally conform to irregular shapes, unlike rigid grippers that require exact positioning.

Consumer Electronics & Assembly

mid-size

Target: Manufacturers handling fragile or variable components

If you are an electronics or consumer goods manufacturer losing product to damage during robotic assembly — this project created multiple enhanced hand prototypes with integrated sensors from 4 different research teams. The soft compliant design means the gripper absorbs minor misalignments and variations in part geometry, reducing reject rates on assembly lines that handle mixed or delicate components.

Frequently asked

Quick answers

What would it cost to implement these soft grippers in our facility?

The project data does not include unit pricing or commercialization costs. However, the core design principle — simplicity and compliance over complex rigid mechanisms — suggests lower hardware costs than traditional multi-fingered robotic hands. Contact the coordinator team for licensing or development partnership pricing.

Can these grippers work at industrial speed and scale?

The project delivered 4 prototypes each from two partner institutions specifically for integration testing at OCADO's warehouse operations, indicating the technology was being prepared for industrial-scale environments. Full production-line speed validation would depend on your specific application and throughput requirements.

Who owns the IP and how can we license it?

The consortium of 7 partners across 5 countries jointly developed the technology under EU funding rules, meaning IP is shared among partners. Key technology holders likely include TU Berlin (coordinator), University of Pisa, DLR, and IIT who all built hand prototypes. Licensing discussions would need to go through the relevant partner holding the specific IP you need.

How robust are these hands in real working conditions?

The project objective specifically targeted robustness and strength as key design goals, addressing the fragility problem in current manipulation systems. The deliverables included 'enhanced hand prototypes' with integrated sensors, and intermediate integration testing with industrial partners OCADO and Disney suggests real-world durability was evaluated.

Do we need to reprogram the gripper for every new product shape?

No — that is the central advantage. The soft hands use what the project calls 'environmental constraints' to naturally adapt their grasp. The compliant material deforms around objects, so the same gripper handles varied shapes without per-object programming. This is fundamentally different from rigid grippers that need exact grasp planning for each item.

What is the current development status — is this ready to deploy?

The project ended in April 2019 with multiple working prototypes delivered and tested in integration scenarios. Based on available project data, the technology reached validated prototype stage with industrial testing at OCADO and Disney. Further engineering would be needed to move from tested prototypes to a deployable product line.

Consortium

Who built it

The SoMa consortium brings together 7 partners from 5 European countries (Austria, Switzerland, Germany, Italy, UK), with a mix of 3 universities, 2 research organizations, and 2 industrial partners making up 29% of the consortium. The coordinator is TU Berlin, a top-tier technical university. The industrial involvement is notable: OCADO, one of the world's largest online grocery platforms, and Disney were recipients of prototype hands for real integration testing — not just advisory board members. The absence of SMEs means no small-company agility in commercialization, but the direct pipeline to major industry users like OCADO provides a credible path from research to deployment. DLR (German Aerospace Center) and IIT (Italian Institute of Technology) add heavyweight robotics expertise.

TECHNISCHE UNIVERSITAT BERLINCoordinator · DE
INSTITUTE OF SCIENCE AND TECHNOLOGY AUSTRIAparticipant · AT
OCADO INNOVATION LIMITEDparticipant · UK
DEUTSCHES ZENTRUM FUR LUFT - UND RAUMFAHRT EVparticipant · DE
UNIVERSITA DI PISAparticipant · IT
FONDAZIONE ISTITUTO ITALIANO DI TECNOLOGIAparticipant · IT
THE WALT DISNEY COMPANY (SWITZERLAND) GMBHparticipant · CH

How to reach the team

The coordinator is Technische Universität Berlin. Use SciTransfer's lookup service to find the project lead's direct contact details.

Order a report on this consortium See TECHNISCHE UNIVERSITAT BERLIN profile →

Next steps

Talk to the team behind this work.

Want to explore whether SoMa's soft gripper technology fits your automation challenge? SciTransfer can connect you directly with the right research partner — contact us for a tailored one-page brief.

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