Modelled field robots could cut organic onion costs by 75%

A 200 kg battery-electric field robot could cut organic onion production costs from €0.52 to €0.13 per kilogram — a 75% reduction — according to modelling published in Nature Communications. Before anyone phones a dealer, note the word "modelling". No such robot has run a commercial hectare.

Researchers at the University of Münster in Germany and the University of California, Davis, assessed the economic and environmental potential of small autonomous robots across onions, lettuce, strawberries and peanuts, under both organic and conventional management. Their hypothetical machine measures 0.85 × 0.70 × 0.70 metres, carries cameras, GPS and lidar, and performs soil preparation, sowing, weeding, irrigation and fertilisation. Harvesting stays conventional, with an allowance built into the cost figures.

Where the savings land

The headline numbers, calculated as production cost per kilogram of harvested crop:

• Organic onions: €0.13/kg robot-based versus €0.52/kg conventional — a 75% saving
• Organic lettuce: €0.70/kg versus €1.32/kg — roughly 47% cheaper
• Organic strawberries: €3.60/kg versus €3.55/kg — the robot actually costs slightly more
• Conventional peanuts: €0.47/kg versus €0.49/kg — effectively a wash

The pattern is blunt. High-value, labour-intensive organic vegetables win; strawberries and peanuts don't. The economics hinge on replacing expensive manual weed control, which is precisely where organic onion and lettuce growers bleed cash.

What does this change for growers today?

Honestly? Nothing yet. These are not measured savings from working farms. The researchers built a theoretical robot and ran engineering calculations, crop-production data and published cost estimates through a model. Two assumptions do most of the lifting: the robot works 18 hours a day, and precision application halves pesticide, fertiliser and irrigation water use compared with conventional practice.

Working widths in the model run from 0.2 to 0.7 metres. That's a very small machine doing very long shifts.

The input assumptions matter more than the hardware price. Variable inputs — seeds, fertilisers, water and labour — account for 72–90% of the modelled production costs. The researchers conclude the economic benefits depend more on cutting these inputs than on the robot's purchase cost. Buy an expensive robot and still spray broadly, and the sums collapse.

The model also skips some inconvenient details. Breakdowns, charging logistics, variable field conditions and external AI and software costs are not fully accounted for. Anyone who has watched a prototype robot discover mud knows why that gap matters.

Emissions and the fine print

The team also ran greenhouse gas calculations, suggesting small electric robots could reduce emissions per kilogram of crop — though the outcome swings hard on crop type, input requirements and yield. Again, modelled, not measured.

The authors are refreshingly candid about all this. They caution that their results are not forecasts of actual farm performance, and they call explicitly for field trials and farm-specific economic assessments to test whether the theoretical advantages survive contact with actual soil.

The verdict

The study's real contribution is framing, not forecasting. It identifies where small-robot economics plausibly work — organic alliums and salad crops, where mechanical weed control replaces costly hand labour — and where they don't. It also tells robot developers that input reduction, not sticker price, decides the business case.

For growers, the watch-point is straightforward: trials of small weeding robots on organic onions and lettuce, with published cost-per-kilogram data from real fields. Until those numbers exist, treat 75% as a hypothesis. A well-built one, but a hypothesis.