What is real now—and what comes next
Commercial laser weed control
Carbon Robotics markets a current LaserWeeder G2 product family using computer vision and high-power lasers for nonchemical weed control. Published product pages identify the systems as Class 4 laser products.
https://carbonrobotics.com/laserweeder-g2-400
Precision agriculture adoption
USDA Economic Research Service reports broad adoption of automated guidance on major U.S. crops, while adoption rates vary significantly by farm size and technology type.
https://ers.usda.gov/publications/105893
Individual-plant robotics
USDA-funded work is pushing robotics, machine vision and AI toward plant-level decision-making, including controlled-environment agriculture and field surveillance.
https://www.ars.usda.gov/research/project?accnNo=448608
Optical pest interception
A 2024 Scientific Reports paper demonstrated a field-scale optical “Photonic Fence” that used machine vision and laser energy to identify and intercept target insects while distinguishing a non-target bee and preventing engagement when people or animals entered the active zone.
https://www.nature.com/articles/s41598-024-57804-6
Edge AI and machine perception
Current agricultural equipment already uses onboard cameras, edge computing and machine learning for real-time crop and weed discrimination. Aurora extends that pattern into a reusable electronics and sensing backbone shared across multiple products rather than one dedicated sprayer or weeding implement.
https://www.deere.com/en-us/our-company/technology-and-innovation/sense-and-act
Large agricultural training datasets
Carbon Robotics states that its current Large Plant Model is trained on 150 million labeled plants across more than 100 crops, demonstrating the scale of field data now feeding agricultural machine vision.
https://carbonrobotics.com/carbon-ai
Neuromorphic agricultural sensing
Recent research has demonstrated spiking-neural-network approaches for low-power adaptive soil-moisture sensing, supporting Aurora's use of neuromorphic computing as an emerging option for selected distributed sensing tasks rather than a blanket replacement for conventional processors.
https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1816292/full
Aurora expansion
Aurora's contribution is the broader systems concept: combine weed control, crop protection, terrain-specialized mobility, small-scale/home systems, plant-level water and nutrient management, and an independent safety architecture rather than treating laser weeding as a single-purpose implement.