🧵 Farming robots are no longer experimental. They're deployed, profitable, and reshaping agriculture. In orchards, vineyards, vegetable fields, and beyond, they're tackling labor shortages, precision spraying, and chemical reduction at scale. This is how robotics is becoming the backbone of next-gen agriculture. [Save this thread for later 📌]

The global farm labor crisis isn’t new, but automation’s response finally is. From autonomous tractors to drone-mounted crop scouts, robots are now performing end-to-end workflows: → Planting → Monitoring → Weeding → Harvesting ... and they’re getting smarter with each season.

John Deere made its biggest robotic play yet: acquiring GUSS Automation, a leader in autonomous sprayers. → 2.6M acres sprayed → 500K operational hours → 90% chemical reduction Precision ag isn’t just about data. It’s about machines that act on it.

Then there’s FarmWise, using AI-powered weeders that operate in the field without supervision. No pesticides. No human fatigue. Just a robot pulling unwanted plants at industrial scale. It’s the Uber for weeding, but built on machine vision and torque.

Seattle-based Carbon Robotics took agricultural robotics a step further with its LaserWeeder G2. It uses deep computer vision, and dozens of CO₂ lasers to identify and vaporize weeds at sub-millimeter precision, processing 4.7M images per hour, zapping up to 600,000 weeds per hour, and leaving soils untouched. Globally, it's deployed over 150 units, killed 30B+ weeds across 100+ crops, and is now scaling globally with modular models.

Startups aren’t standing still. → Bonsai Robotics raised $15M to build autonomous nut harvesters → Designed for dusty orchards with minimal visibility → Trained on LIDAR + edge AI Goal: replace seasonal labor with full-time automation, no GPS needed. It's me visiting them (farm-ng n back then) last year.

In Australia, SwarmFarm Robotics is building an open platform for modular, right-to-repair farm bots. Each robot, named “Robbie” or “Bottomley Potts”, can spray, seed, or cultivate. Farming becomes scalable, decentralized, and resilient. Their slogan? “Let robots do the dirty work.”

This is a fully electric, driverless tractor that handles ploughing, seeding, spraying and harvesting by itself. It’s already working fields in Hebei Province, and a nationwide rollout is planned as China deals with an aging rural workforce and labor shortages. A single farmer running a small robot fleet could free up millions of people for other jobs and cut China’s reliance on Western machinery.

Why this matters: → $30B+ global ag labor shortage → Stricter environmental regs → Pressure for local, sustainable production Ag robotics isn’t a moonshot. It’s an infrastructure layer, like tractors were in 1910.

The future of farming won’t look like Wall-E. It’ll look like swarms of compact robots, working soil gently, responding in real time, and learning from each harvest. If you enjoy my work in physical AI space, please subscribe to my robotics newsletter:

Can we just shake the tree

Spot on! The shift from experimental tech to real commercial profitability is happening much faster than expected. Solving labor shortages while increasing precision in orchards and vineyards is a massive win for sustainable farming. The future of agriculture is definitely autonomous

I feel like I've seen these before...

That's Tevel's flying robots. What most people miss is the logistics: I'd bet the ground carts collecting from multiple drones in parallel are the real throughput enabler, not the picking itself.

Have you ever seen how fast temporary workers operate? It's an unreal pace that makes this hilarious as of now. If you can speed that up like 10x then you might have something solid. Consider an arm system with a roller vs drones that moves really fast. Readily available.

Agriculture is a compelling robotics market because the tasks are repetitive but the environment is messy. I’d be interested in annual utilization: a machine that works across crops and seasons has a very different payback than one built for a short harvest window. Which use case has the clearest economics?

impressive how fast robots moved from labs to profit

Moving from spraying/weeding to actual delicate harvesting (like the apple drones shown here) has always been the hardest leap. Vision models + soft grippers finally reaching commercial speed without bruising the produce is a massive milestone.

@grok 以影片中的使用範例，評估需要多少台機器才能與一個專業的採收工人速度相比？成本為多少？延伸的其它支出項目為何？

Shit this looks cool AF

looks like a waste of power. and loud. Why not use hydrogen balloons and pump water into ballasts to control height? basically opposite of sub marine. Make a super marine picker.

The real milestone is unit economics, not a flashy demo. If these systems reduce chemicals and survive harvest-season uptime, adoption can move very quickly.

Lukas, I really enjoy your posts, but there is no way this drone apple picking system is a financially viable solution. Cycle time alone this seems dead on arrival. Maybe some of the others on your list...

Wow. Thats amazing. Is there a good resource to keep up on all agricultural robot developments?