FarmBot

[[farmbot-genesis|FarmBot]] is the most accessible point of entry into modern agricultural robotics. The hardware is small enough to mount over a 3×6-foot raised bed in a residential backyard. The full design files, parts lists, assembly instructions, and software source code are publicly available and licensed for free reuse and modification. As of the mid-2020s, several thousand FarmBot units operate in homes, schools, research labs, and small farms across more than 100 countries [1, 2].

For the [[Vault Source Roadmap|source roadmap]]‘s Tier 1 agricultural-robotics gap, FarmBot is the strongest single anchor. The project is mature, well-documented, [[mission-district-sf|mission]]-aligned, and operational at the scale 0mn1.one’s hemp-and-garden-and-restaurant vision actually contemplates — not a 5,000-acre commodity farm, but small-to-medium intensive plots. The commodity-ag robotics world (John Deere autonomy, Carbon Robotics weed-zapping lasers, Iron Ox enclosed-warehouse hydroponics) is also real and is documented elsewhere; FarmBot is the canonical small-scale open-source case.

What it is

The basic system [1, 2]:

• Gantry-style robot mounted over a rectangular raised bed. Two parallel rails (the X-axis) carry a moving cross-rail (the Y-axis), which carries a Z-axis tool head. The combined motion can position the tool head anywhere over the bed with sub-millimeter precision.
• Tool head with interchangeable tools — seed injector, watering nozzle, weeding knife, soil-moisture probe, camera. Tool changes are automatic via a magnetic-mount tool rack at one end of the bed.
• Water and power routed along the gantry. Standard 24V DC power from a wall transformer; standard garden-hose water connection. Optional on-bed solar power for [[off-grid-living|off-grid]] operation.
• Software stack running on a [[raspberry-pi|Raspberry Pi]] attached to the gantry. Web-based UI accessible from any browser. Users define their bed layout, plant types, planting calendar, and care schedules; the [[farmbot-genesis|FarmBot]] executes accordingly.
• Cloud sync (optional) for software updates, crop-image archive, and remote management. Users can also self-host [[the-cloud-of-unknowing|the cloud]] component.

Three principal hardware lines [1]:

• Genesis — the original line, 1.5m × 3m bed. Most popular for residential and small-farm use. Approximately $3,000 USD as of 2026.
• Genesis XL — 3m × 6m bed. Targeted at small commercial farms, schools, research labs. Approximately $4,500 USD.
• Express — a smaller, lower-cost line designed for educational use. 1.2m × 1.2m bed. Approximately $1,500 USD.

Compared to commodity agricultural robotics, [[farmbot-genesis|FarmBot]] is small. Compared to a backyard gardener doing the work by hand, [[farmbot-genesis|FarmBot]] is automation. The fit is for the in-between — intensive small-scale food production where precision matters per plant, not per acre.

What it does

The current capabilities [1, 2]:

• Seeding. The robot picks up seeds from a tray, injects them at the right depth, at the right spacing, on a defined planting schedule. Multiple seed types per bed.
• Watering. Per-plant watering rather than blanket irrigation. A drip-style nozzle delivers a metered amount to each plant at programmed intervals. Water-conservation gains over conventional irrigation are real.
• Weeding. A weeding knife that descends to a precise position and rotates can remove weeds between cultivated plants. Image recognition identifies what is weed versus crop; precision is good for established plants, less reliable for very young seedlings.
• Soil monitoring. Soil-moisture and soil-temperature probes can be deployed across the bed; data feeds into the watering schedule.
• Imaging. A downward-facing camera takes regular images of each plant; users can review crop progress remotely; the system builds a per-plant growth history.
• Custom tools. The tool-mount system is documented and user-extensible. Several community-developed tools exist (mulch-applicator, fertilizer-injector, more sophisticated weeders).

The capabilities are real but bounded. [[farmbot-genesis|FarmBot]] does not harvest; it does not transplant; it does not work outside the rectangular gantry’s reach; it does not deal with anything taller than the gantry clearance (so corn, sunflowers, and tomato cages need manual management). It is precision automation for a particular slice of the planting-tending workflow, not a fully autonomous farm.

What FarmBot demonstrates

For 0mn1.one’s [[mission-district-sf|mission]], [[farmbot-genesis|FarmBot]] is significant beyond its specific capabilities [1, 3]:

• Precision agriculture is feasible at residential scale. The technical components — [[stepper-motor|stepper]] motors, microcontrollers, computer vision, programmable irrigation — are now cheap and mature enough that a small open-source team can integrate them into a working product. The barrier to small-scale agricultural automation is no longer fundamental engineering; it is integration, support, and labor cost.
• Open-source hardware is a viable model in agriculture. [[farmbot-genesis|FarmBot]]‘s design files are public; users can assemble their own from raw parts (cheaper but more labor) or buy pre-fabricated kits (more expensive but faster). The community has produced documentation, modifications, and translations in dozens of languages. The open-source model has not displaced proprietary agricultural automation, but it has demonstrated that the model works at this scale.
• Per-plant management is a different paradigm from per-acre management. [[industrial-agriculture|Industrial agriculture]] optimizes per-acre — same treatment to all plants, with the average plant doing the work. FarmBot makes per-plant treatment cheap. Each plant can have its own watering schedule, its own pest-monitoring regime, its own data history. Polyculture beds — corn-beans-squash, [[three-sisters|three sisters]], herb gardens — become tractable when the management cost per plant is low.
• The data substrate matters. FarmBot’s cloud component (and its self-hosted alternative) accumulates a per-plant, per-bed, per-season record. The data is the grower’s; over years, a FarmBot installation builds a knowledge base about that specific bed and that specific gardener’s preferences. This is one of the substrates a serious small-scale agriculture practice would use to compound knowledge.
• The labor reframing. FarmBot does not replace the grower; it does the repetitive precision work and frees the grower to do the higher-value activities (planning, harvest, observation, judgment). This is the augmentation pattern [[doug-engelbart|Engelbart]] named — augmentation rather than automation — applied to agricultural labor.

Limits worth naming

FarmBot is not the right tool for several things 0mn1.one’s [[mission-district-sf|mission]] would also need [1]:

• It does not scale to acres. A 3×6m bed is not a hemp field. Different robots are needed for different scales; FarmBot occupies one slice of the spectrum.
• It is not the right tool for taller perennial crops (orchards, agroforestry, vineyards) — the gantry geometry assumes ground-level annuals.
• It is not the right tool for livestock or animal management — that is a separate robotics domain (Burro, autonomous sheep dogs, robotic dairy systems).
• It is not designed for harvesting — the harvest step in horticulture is still the most labor-intensive and least automated. Harvesting robots are an active research area; mostly not yet at FarmBot’s level of accessibility.
• It still requires a grower. Setup, maintenance, plant selection, problem diagnosis, harvest, and integration with the rest of the operation all require human attention. The robot saves labor; it does not eliminate it.

A serious agricultural-robotics strategy for 0mn1.one’s [[mission-district-sf|mission]] would use FarmBot as one anchor among several — small-scale gantry robots for intensive beds, mobile small robots like Burro or Naïo for between-bed tasks, larger commodity-scale autonomy for any acreage scale operations, and harvest-specific robotics as that field matures. None of these substitute for the grower; all of them shift the labor profile.

Cultural and historical

FarmBot was started in 2011 by Rory Aronson, then a mechanical engineering student at Cal Poly San Luis Obispo, who built the first prototype as an undergraduate project [1]. The project went public in 2014 with a viral whitepaper laying out the open-source agricultural-robotics vision; FarmBot Inc. was incorporated in 2016 to handle the production and support of pre-built kits.

The community grew rapidly through the mid-2010s. The official forum, GitHub repositories, and language-localized documentation have been active continuously since. FarmBot is now used in multiple agricultural-research programs, hundreds of schools (including a sustained presence in Maker and STEM curricula), and a dispersed network of residential and small-farm users worldwide.

The project sits in a broader open-source agricultural-technology tradition — alongside Open Source Ecology (Marcin Jakubowski’s Global Village Construction Set), Sensorica, FarmHack, and the [[open-source-seed-initiative|Open Source Seed Initiative]]. None of these have displaced proprietary agricultural systems; collectively they are evidence that an alternative substrate exists and is functioning.

Why this matters for 0mn1.one

The hemp-homes pillar has [[hempcrete]] and [[hemp]] as its substrate; the autonomous-farms pillar needs equivalent substrate. FarmBot is the first wiki entry on agricultural robotics; it begins the substrate. The platform’s future entries on Burro, Iron Ox, Naïo, Carbon Robotics, John Deere autonomy, and harvest-specific systems will fill the rest.

The project also demonstrates the kind of venture [[0mn1one|the platform]]‘s [[mission-district-sf|mission]] would look like in physical-infrastructure form. Open-source. Mission-aligned. Patient capital (FarmBot has been profitable but has never raised the venture rounds that would have demanded faster scaling). Community-supported. Documentation-rich. Operating at the scale of small farms and intensive gardens rather than commodity acreage. If 0mn1.one ever ships hardware — gardening robots, hemp-processing equipment, geodome controllers — the FarmBot model is a viable shape to consider.

Lenses still to grow

• The full hardware specification — the bill of materials, the gantry design, the tool-mount system, in detail; specific information for someone considering self-build vs. kit-build
• The software stack — FarmBot OS, the web app, the API, the relationship to [[raspberry-pi|Raspberry Pi]] and other off-the-shelf substrates
• Specific user case studies — schools using FarmBot in curriculum, small farms integrating it into operations, research applications
• The economics in detail — kit costs, self-build costs, time savings per square meter, comparison with manual labor and with proprietary alternatives
• The community — forums, GitHub activity, language-localization efforts, the modification ecosystem
• Comparison with adjacent open-source ag projects — Open Source Ecology, FarmHack, Sensorica, the Open Source Seed Initiative
• Comparison with proprietary small-scale ag robotics — Iron Ox, Naïo, the Burro mobile platform; what each does that FarmBot doesn’t, and vice versa

See also

Auto-generated from this entry’s typed relations: frontmatter, grouped by relation type so the editorial signal isn’t flattened.

• Enables: [[0mn1one]]
• Shares approach with: [[permaculture]] · [[regenerative-finance]]

Sources

1. FarmBot Inc. Public documentation, hardware design files, software source code, and assembly guides at farm.bot. Self-published primary source.

2. FarmBot whitepaper (Aronson, 2014). The original public proposal that launched the project; freely available. Worth ingesting in primary form.

3. Wikipedia: FarmBot. Brief biographical and project history. https://en.wikipedia.org/wiki/FarmBot

This entry currently rests on the project’s public documentation and widely-known characterization. Future revisions could deepen with primary-source quotation from the 2014 whitepaper, specific user case studies, and a fuller technical architecture section. The [[Vault Source Roadmap|vault source roadmap]] flags agricultural robotics broadly as a Tier 1 gap; this entry begins to close it.

Page filed 2026-05-03. First wiki entry on agricultural robotics; expect companion entries on Burro, Iron Ox, Naïo, Carbon Robotics, and John Deere autonomy in subsequent passes.