The American agricultural startup has developed a machine capable of eliminating up to 600,000 weeds per hour with sub-millimeter precision, without the need for herbicides.
Agriculture is undergoing a true technological revolution. In response to increasing weed resistance to herbicides and the global labor shortage in the field, robotics and artificial intelligence have become key allies. A prime example is Carbon Robotics, founded in Seattle in 2018, which has unveiled its new LaserWeeder G2, a machine that looks straight out of science fiction: 24 high-power lasers and 24 NVIDIA GPUs working in parallel to destroy up to 10,000 weeds per minute.
600,000 weeds per hour, chemical-free
The LaserWeeder G2 can vaporize approximately 600,000 weeds per hour, which equates to 167 per second. Its precision is sub-millimeter, ensuring only weeds are removed while crops remain untouched.
This level of accuracy is achieved thanks to the Large Plant Model, an AI model trained with over 65 million agricultural images that reliably identifies each plant, even under complex lighting, humidity, or terrain conditions.
The company claims that this machine can replace the work of 75 people engaged in manual weeding, representing significant cost and time savings for farmers.
A solution to herbicide resistance and labor shortages
The intensive use of herbicides over recent decades has led many weed species to become resistant, forcing an increase in chemical doses and risking both soil health and environmental safety. The LaserWeeder G2 offers a radically different alternative: eliminating weeds without chemicals, solely with AI-guided laser energy.
Additionally, Carbon Robotics is exploring full automation of agricultural machinery through retrofitted autonomous tractors. In this model, a single farmer could remotely control multiple vehicles, alleviating the tractor driver shortage affecting countries like the United States.
Benefits, risks, and challenges
The advantages of this technology are clear:
- Reduced chemical use: promotes more sustainable farming.
- Labor cost savings: replaces entire crews in a context of labor shortages.
- Higher precision: less collateral damage to crops.
- Agricultural data: each machine also functions as a data collector to enhance predictive models.
But there are also challenges:
- High cost: using 24 NVIDIA GPUs and high-power laser technology drives up the machine’s price.
- Safety: laser use poses risks to workers (eyes, skin) and, in arid zones, potential fires.
- Root persistence: if a root isn’t completely destroyed, the weed could regrow.
NVIDIA, a strategic partner
The collaboration with NVIDIA isn’t coincidental. The company’s GPUs not only provide the processing power needed for real-time data processing onboard the machine but are also used later in Carbon Robotics’ data centers to train the next generation of agricultural AI models.
This synergy of hardware, AI, and robotics reinforces NVIDIA’s leadership in a burgeoning market: AI applied to agriculture, a sector moving hundreds of billions of dollars and urgently in need of more efficient and sustainable solutions.
Future outlook
To date, Carbon Robotics has sold over 150 units of its LaserWeeder in 14 countries, eliminating more than 30 billion weeds. Its goal is clear: to establish itself as a global leader in chemical-free precision agriculture.
The next step is to scale its technology to autonomous tractors and intelligent farming platforms, capable of operating 24/7 under remote supervision.
As CTO Alex Sergeev states: “There is no weed resistant to lasers.”
Frequently Asked Questions (FAQ)
1. How much does the LaserWeeder G2 cost?
Carbon Robotics does not publish official prices, but analysts estimate that the cost could surpass that of traditional agricultural machinery by several hundred thousand dollars, owing to the intensive use of GPUs and laser technology.
2. Will it completely replace herbicide use?
In many crops, yes, although in some cases farmers will continue to combine chemical, mechanical, and now laser control techniques.
3. Is it safe to use lasers in an agricultural environment?
The machine includes safety protocols, but there is always a risk of injury or fires. Therefore, it is recommended to use it under controlled conditions.
4. What role does NVIDIA play in this project?
NVIDIA provides the necessary computing power both in the field (real-time processing) and in data centers for training AI models.