SWEEPER · Project

Autonomous Robot That Picks Sweet Peppers in Greenhouses Without Human Labour

foodPilotedTRL 7

Imagine a robot arm that can reach into a greenhouse full of pepper plants, spot the ripe ones among all the leaves, grab them gently, and cut them off — all on its own. That's what SWEEPER built. It started from an earlier research prototype and turned it into something growers could actually watch working in their greenhouses. The team also figured out how to arrange the plants so the robot can reach the peppers more easily, and they swapped a complex 9-joint arm for a simpler 4-joint one to cut costs dramatically.

By the numbers

4DOF

Simplified robot arm (down from 9DOF), greatly reducing costs

Consortium partners across 4 countries

64%

Industry partners in consortium

TRL 6

Starting technology readiness level from CROPS project

Total project deliverables completed

SME partners in the consortium

The business problem

What needed solving

Finding enough workers to hand-pick sweet peppers in hot, humid greenhouses is getting harder and more expensive every year across Europe. The repetitive, physically demanding work in harsh conditions drives workers away, and growers face rising labour costs that squeeze their margins and threaten the competitiveness of European greenhouse horticulture. Automation is the obvious answer, but until recently no robot could reliably spot, reach, and gently pick peppers without damaging them.

The solution

What was built

SWEEPER built a working sweet pepper harvesting robot with a patent-pending gripper that can grasp peppers without knowing their exact position, a simplified 4DOF robot arm (down from 9DOF to cut costs), model-based computer vision for detecting fruit among dense foliage, and a LightField sensor capturing both colour and 3D information simultaneously. The team also developed optimized cropping systems that make plants easier for robots to harvest. All 36 deliverables were completed, including live demonstrations to growers.

Audience

Who needs this

Large greenhouse operators growing sweet peppers or high-value vegetablesAgricultural robotics companies looking to license or co-develop harvesting technologyFresh produce supply chain companies facing labour-cost volatilityGreenhouse equipment manufacturers wanting to add automation productsAgricultural cooperatives in labour-scarce regions of Europe

Business applications

Who can put this to work

Greenhouse Horticulture

mid-size

Target: Large-scale greenhouse operators growing sweet peppers or similar high-value crops

If you are a greenhouse operator struggling to find workers willing to pick peppers in hot, humid conditions day after day — this project developed an autonomous harvesting robot with a patent-pending gripper that can locate and pick sweet peppers without needing to know their exact position. It was demonstrated to real growers and uses a simplified 4-degree-of-freedom arm instead of 9, which greatly reduces equipment costs.

Agricultural Robotics & Equipment

SME

Target: Companies manufacturing or integrating agricultural automation equipment

If you are an ag-tech equipment maker looking to add harvesting robots to your product line — SWEEPER produced a working system built on the open-source Robot Operating System (ROS) with tested modules for vision, gripping, and arm control. The 4DOF arm design and patent-pending gripper are ready for licensing or co-development, backed by real greenhouse demonstration data from an 11-partner consortium across 4 countries.

Fresh Produce & Food Processing

enterprise

Target: Fresh produce packers and vegetable supply chain companies

If you are a fresh produce company facing rising procurement costs because growers cannot find enough harvest labour — this project proved that robotic harvesting of sweet peppers is economically and technically viable. Automating the harvest could stabilize your supply and reduce the labour-cost volatility that drives up prices, especially for high-value crops where manual picking is the biggest expense.

Frequently asked

Quick answers

What would a harvesting robot like this cost compared to manual labour?

The project specifically redesigned the robot arm from 9 degrees of freedom down to 4, which the team says greatly reduces costs. While exact pricing is not published, the objective states the robot was shown to be economically viable for sweet pepper harvesting. The cost advantage grows as labour shortages push wages higher.

Can this work at industrial scale across a full greenhouse?

SWEEPER conducted live demonstrations to growers in real greenhouse conditions. The system was built on proven hardware and software modules that started at TRL 6. However, scaling to full commercial deployment across multi-hectare operations would likely require further engineering for speed and reliability.

What about intellectual property and licensing?

The project includes a patent-pending gripper end-effector that can grasp sweet peppers without needing an accurate measurement of the fruit's exact position and orientation. The software is built on the open-source Robot Operating System (ROS). Contact the coordinator at Wageningen Research to discuss licensing terms for the patented components.

Does the robot work with existing greenhouse setups?

Not necessarily out of the box. SWEEPER found that different growers use different cropping systems with varying crop density, and the project specifically optimized the cropping system to facilitate robotic harvesting. Adopting this technology may require adjusting how you grow and train your plants.

How mature is this technology — is it ready to buy?

The project started from TRL 6 components and was funded as an Innovation Action, meaning it was focused on bringing technology closer to market. Live demonstrations were conducted for growers with 36 deliverables completed. It is past the prototype stage but likely needs a commercial partner to turn it into a purchasable product.

Can this robot harvest other crops besides sweet peppers?

SWEEPER was specifically designed and optimized for sweet peppers. The vision system, gripper, and crop optimization are all tailored to this crop. Based on available project data, adaptation to other high-value greenhouse crops would require additional development, though the underlying ROS platform and 4DOF arm design could serve as a foundation.

What regulations apply to deploying robots in greenhouses?

Based on available project data, the project focused on technical development and grower demonstrations rather than regulatory certification. Standard machinery safety directives (CE marking) and workplace safety regulations would apply. The consortium included 7 industry partners which suggests awareness of commercialization requirements.

Consortium

Who built it

The SWEEPER consortium of 11 partners across 4 countries (Belgium, Israel, Netherlands, Sweden) is heavily tilted toward industry with 7 industrial partners making up 64% of the team — a strong signal that this was built for real-world use, not just academic publishing. Wageningen Research, one of the world's top agricultural research institutions, leads the project. The inclusion of 3 SMEs alongside larger industry players suggests both entrepreneurial agility and established distribution channels. With only 2 universities involved, the balance clearly favours applied development over basic research. The international mix brings together Dutch greenhouse expertise, Israeli precision agriculture know-how, and Swedish robotics engineering.

STICHTING WAGENINGEN RESEARCHCoordinator · NL
PROEFSTATION VOOR DE GROENTETEELTparticipant · BE
BEN-GURION UNIVERSITY OF THE NEGEVparticipant · IL
B&A AUTOMATION BVBAparticipant · BE
UMEA UNIVERSITETparticipant · SE

How to reach the team

Wageningen Research (Netherlands) — search for SWEEPER project coordinator at Wageningen University & Research for contact details

Order a report on this consortium See STICHTING WAGENINGEN RESEARCH profile →

Next steps

Talk to the team behind this work.

Want to connect with the SWEEPER team about licensing the harvesting robot or gripper technology? SciTransfer can arrange an introduction and help structure a technology transfer discussion.

Order a report on this topic More in Food & Agriculture