Loading video...

Video Failed to Load

Go Home

Carbon Robotics' Autonomous Weeder can kill over 100,000 weeds/hour using 8 simultaneously operating laser modules that deliver quick zaps on emerging weeds. [video: [read more:

1,604,473 views • 3 years ago •via X (Twitter)

10 Comments

Cory Gabrielsen's profile picture
Cory Gabrielsen3 years ago

I wrote the original laser-shooting algorithm and original self-driving algorithm for this. :)

Duane O’Keefe's profile picture
Duane O’Keefe3 years ago

Rather have a weed laser than weed sprayer…Bayer AG is going to be pissed and possibly try to suppress this tech or purchase it.

Tarak's profile picture
Tarak3 years ago

If machines can do this to plants, machines can also be developed to do this to humans.

JohnInGreenwood's profile picture
JohnInGreenwood3 years ago

Bravo. Fewer chemicals to control weeds! I'm sure Monsanto will do their best to keep this from taking off. I like it - fits perfectly with the electrified economy.

James CryptoGuru's profile picture
James CryptoGuru3 years ago

Imagine an airborn drone version of this that could clean entire fields.

Yaman Tasdivar's profile picture
Yaman Tasdivar3 years ago

Biological NNs have gotten so efficient at quick-zapping unwanted life that is simply trying to survive - but causing a mere inconvenience. On an unrelated note, what’s 8 billion divided by 100,000 per hour? @a_meta4 @TeslaTrip

Pablo Mendoza's profile picture
Pablo Mendoza3 years ago

The Autonomous Weeder may save time and labor, but what about the human connection with nature? Aviators may help you weed out superficial solutions.

Dillon Lyrek's profile picture
Dillon Lyrek3 years ago

No more poison.

Joey Wong王祖兒's profile picture
Joey Wong王祖兒3 years ago

Pls write one for mosquito

Mr Pablo's profile picture
Mr Pablo3 years ago

Weeds beware, technology has your number! Keep your garden pristine with the Autonomous Weeder, the future of gardening is looking bright!

Related Videos

"bUt wHo iS gOIng to pICk tHe cROps"? Want to eat food that hasn't been endlessly doused in chemicals, but which also isn't super expensive? Think there might be an alternative to allowing millions of illegals into the country to harvest the crops? "Precision Agriculture" is finally here, and can deliver that, driving down the cost of groceries without the old inputs of factory farming Imagine a world where agricultural scale doesn't mean vast reservoirs of chemicals getting constantly dumped over the food supply, but high-tech, precision equipment doing what the chemical inputs used to The coolest example is the weeder that zaps and burns thousands of weeds a minute, up to 80 a second, saving farmers money and obviating RoundUp, or other chemicals like it. The combination of AI, precision machinery, and agricultural advancement can make that happen, the world it could create is a really exciting one What makes it all the more so is that these innovations aren't just cool. They're real leaps forward that will help Americans. They'll lower the cost of groceries, help farmers avoid huge input bills, such as for the herbicides obviated by the laser weeder, and create a thriving domestic manufacturing sector built around the production of this precision machinery, and all the advances that go into creating it. The pairing of technology and manufacturing in a really cool way that we haven't really seen outside of the military-industrial complex for decades.That's finally possible again, and Precision Agriculture looks like it can make it so

Jason Robertson

39,618 views • 6 months ago

JUST IN: Dyna Robotics just published one of the most important research papers in robotics this year. It could fundamentally change how robot foundation models are trained. A scaling law that transfers from human video to robot performance. Dyna-2 is out and it's 🔥 Here's what that means in plain terms. Dyna-2 was pre-trained on ONE MILLION hours of egocentric human video, 170 years of continuous human experience, cooking, folding, assembling, cleaning. And as that human data scaled, robot performance improved. Predictably. Monotonically. Across 39 tasks on two different robot embodiments the model had never seen. → 1,000 hours pre-training → 20% normalised task performance → 10,000 hours → 28% → 100,000 hours → 45% → 1,000,000 hours → 53% Human video exists at effectively unlimited scale. Every cook, every factory worker, every craftsperson wearing a camera is generating training data for future robots. But the finding that stunned even the researchers, world modeling is what makes the transfer work. A model trained to predict future video AND actions massively outperforms one trained on actions alone. Video is the new scaling axis for robotics. One more jaw-dropping data point. 13 minutes of teleoperation data was enough to fine-tune Dyna-2 to open a bottle cap using two five-fingered robot hands. The robots are coming, and they're learning from us directly :D Read more here: Congrats Jason Ma and team! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

23,373 views • 16 days ago