How far the humble GPU has come, from mere 3D accelerator to integral component of global billion-dollar industries. Advancements in computational power naturally lend themselves to improvements with robotics, with one developer catching my eye as they leverage the power of Nvidia GPUs to rapidly wipe out weeds using lasers.
Meet Carbon Robotics and its LaserWeeder G2 Series. This farming equipment consists of modules that each pack two 240W lasers, three high-resolution cameras, 20 LED lights, and two Nvidia GPUs. The largest configuration the company offers, G2 1200, totals at 16 modules, and weighs a staggering 8,165kg factoring in the enormous water cooling systems that come part and parcel.
Using a ‘Large Plant Model’, trained on a dataset of over 65 million images, this hardware comes together to provide a more plant-friendly alternative to chemical weed control. As Carbon Robotics’ chief technology officer, Alex Sergeev, puts it: “There’s no such thing as a laser-resistant weed.” Anyone else feel like they’ve entered a sci-fi flick from the ’80s?
It’s hard not to think of LaserWeeder G2 as anything but a terminator of bothersome plants, as the unit can incinerate up to 10,000 weeds per minute. That’s over 100 per second. Looking out at my garden, I certainly wish I could put my GeForce RTX 5090 to work in such a fashion, providing I could trade a tractor for a more Roomba-sized device.
Carbon Robotics has shipped 150 LaserWeeders across the globe since the company’s inception in 2018. That’s just over 21 super-smart laser-incinerating systems a year. I know I’ll be keeping my eye out for one as I pass by fields on the train or during my next road trip.
While I wish Nvidia or Carbon Robotics were more transparent about what Nvidia GPUs are powering these weed killers, it doesn’t nullify any of the nerdy glee I get from seeing them in action. While GeForce is something of an afterthought for Team Green these days, financially speaking, it does make me smile knowing such advancements sprouted from the seed of gaming.
If you’re curious to learn more about how other developers are using GPUs in robotics, check the source link down below. Otherwise, be sure to catch up on all the latest Club386 coverage via our Google News page and socials.
