Nanovel: EIC Accelerator Bet on Robotic Citrus Harvesting

Nanovel says seasonal hand picking accounts for about half of what it costs to produce fresh citrus, and that the fruit can sit up to 1.5 metres inside a dense canopy. The company, based in the Mevo Carmel industrial park in northern Israel, builds a robot that reaches into that foliage with several telescopic arms, holds each fruit with a vacuum gripper and cuts the stem. The EIC Accelerator selected its project at the 1 October 2025 cut-off for blended finance, and CORDIS records an EU contribution of €2,499,999 for a project running from 1 March 2026 to 29 February 2028, signed on 5 June 2026.

The project is eight months old. What it promised can be set against what is already public: field trials in Israel and California, a funding agreement with California's Citrus Research Board, and a set of cost and productivity figures that come from the company and have not been published as trial data.

How the robot picks

The machine is a towed platform carrying several telescopic arms. Each arm carries what the company calls a patented end-effector: a vacuum gripper that cradles the fruit, then cutters that trim the stem, after which conveyor belts carry the fruit to the harvest bins, as described by Fruitnet and Citrus Industry in September 2024. Vision systems and an on-board computer locate the fruit in the canopy and divide it between the arms. For the fresh market, AgFunder News reported in August 2026 that the stem is cut less than 2 mm from the fruit, and that the vision system selects fruit by colour and size but does not yet detect defects.

The number of arms has grown since 2023. In October 2023 the Jerusalem Post described up to four arms, with a plan to double that. A year later Future Farming reported six arms in a modular design that takes four to eight, picking about one 400 kg bin an hour and up to 20 bins in a full working day, at heights up to 3 metres. AgFunder News still describes six arms in 2026. The trial machines are towed and powered by a tractor; both 2024 and 2026 reports describe a self-driving version as in development.

The machine competes with the picking crew, and Mazor is explicit that the robot does not replace the crew entirely. "We're not offering a labor free solution," he told AgFunder News: the robots take roughly 70 to 80% of the fruit and an operator picks what they cannot reach. The CORDIS summary gives a different figure, 90 to 95% removal efficiency. Neither source explains the difference.

What has been tested

The outside partner on record is in California. In September 2024 the state's Citrus Research Board, an industry-funded body, signed a $900,000 non-dilutive agreement with Nanovel for a series of field trials with Californian growers, reported by Fruitnet, Citrus Industry and Israel21c. Citrus Industry wrote that the agreement includes delivery of a robot after the trials, and Israel21c that the board would adopt the robot if the trials succeeded. The trials took place in autumn 2025: Calcalist and Citrus Industry report that validation was completed, and AgFunder News describes a seven-week trial at a leading California grower with a tractor-pulled machine.

A search on 4 November 2026 found no trial results published by the company or the board, and no report of a robot being delivered to the board. The productivity and cost figures in the public record come from the company: in its project summary in CORDIS and in interviews.

What the EIC is paying for

The CORDIS project sets out the plan: take the robot from technology readiness level 6 to 8 through field demonstrations in Spain, Israel and Italy, with a second-generation machine that picks 1.68 bins an hour against 1.26 for the first, running up to 16 hours a day across a harvest season of more than nine months. It names three pilot partners as confirmed: Bollo Natural Fruit in Spain, Mehadrin in Israel and O.P. Coppi in Italy. A search of trade press on 4 November 2026 found no report from any of the three about the trials. The commercial target is 200 second-generation robots in the field by 2030, with projected revenue of €57.9 million. CORDIS gives no total project cost and, eight months in, lists no deliverables or reports.

1. Working multi-arm harvesterExistsSix-arm, tractor-towed machine described by Future Farming in 2024 and AgFunder News in 2026.
2. Field validation in CaliforniaRun, results not publicCompleted in autumn 2025 with the Citrus Research Board, per Calcalist; no trial data published.
3. Pilots in Spain, Israel and Italy, TRL 8In progressProject runs to February 2028; partners named only in CORDIS.
4. Gen-2 at 1.68 bins an hourClaimFigure from the CORDIS summary, not independently measured.
5. Self-driving platformIn developmentPlanned since 2024; still described as in development in August 2026.

The equity has not been announced as closed

The award is blended finance, and no investment from the equity part has been announced. Nanovel announced in July 2026 that the EIC had made it eligible for up to €4.3 million of equity as the anchor of a planned round of about €8 million, reported by Ynet, Calcalist, which calls the company "a candidate" for the investment, and AgNavigator. The dollar figure for the round differs between reports, $9.1 million in Calcalist and FreshPlaza, $9.3 million in Ynet. The EIC Fund portfolio, checked page by page on 4 November 2026, does not list Nanovel, and no closing of the round has been announced.

Mazor funded it himself for its first two years before an undisclosed seed round, according to the Jerusalem Post, and told AgFunder News that it had raised under $10 million to date. No investor has been named in any source found. The register entry with the Israeli Corporations Authority shows the company as active, with its last annual report filed for 2025, and no violation flag; it publishes no accounts or staff numbers.

The market and the timing

The company's own case is labour: seasonal hand picking makes up roughly half of fresh-fruit production costs, a figure in the CORDIS summary and in the July 2026 coverage. Its market estimates vary by source. Calcalist and AgNavigator report about $1 billion a year worldwide for oranges and lemons; AgFunder News reports about $1.5 billion in the company's target regions. The near-term markets are the United States and southern Europe, with Spain and Italy named for trials. The machine is to be sold outright, at about $400,000 each with payback in under five years, through local dealers who also service it, per AgFunder News; Future Farming reported in 2024 that pricing had not been released.

On timing, the July 2026 announcement gives a commercial launch for oranges and lemons in 2028. The AgFunder interview a month later sets out a slower sequence: trials with southern European growers in 2027, pilot deployments in 2028 and commercial deployment in 2029. The EIC project itself ends in February 2028.

An Israeli project funded from Horizon Europe

Israeli companies can take part in Horizon Europe because Israel is associated with the programme, under an agreement signed on 6 December 2021, according to the European Commission and the EU delegation to Israel. The European part of the project is in its pilots: two of the three named partners are in Spain and Italy, and AgNavigator notes that Spain is one of the world's largest citrus producers. Whether European growers buy the machine, and where it would be built, is open: AgFunder News reports that Nanovel is considering manufacturing and assembly partners in California, Spain, Italy, Taiwan or Australia.

The reporting is based on publicly available sources, listed below.