Carbon Robotics Develops Thinning Laser Technology to Curb Dependency on Herbicides

27-02-2023 | By Robin Mitchell

Engineers and scientists are developing tech-driven solutions like robotics and AI to reduce herbicide dependency for weeding in the agriculture industry. However, current systems face challenges with thinning, a crucial step in farming. To address this, researchers are exploring laser-based plant thinning techniques and using cameras and imaging for accurate weed detection. These advancements not only aid in herbicide reduction but also improve the overall efficiency of the agriculture industry. What is thinning, what has Carbon Robotics developed, and how does this further help eliminate humanity’s dependency on herbicides?

What is thinning?

 Agriculture is arguably humanity’s oldest industry, going back thousands of years. One of the major benefits that agriculture allowed was predictable food sources and long-term food storage, which enabled humans to spend more time developing new inventions and making new discoveries. In fact, the importance of agriculture is so critical in technological development that every significant advance in human civilisation has been sparked by an abundance of food as a result of improvements in farming. 

For example, the plough enabled humans to tile the earth and make better farming conditions for crops, the domestication of animals helped massively with food and labour, and the development of artificially sourced fertilisers eliminated the need for farmers to try and regenerate soils naturally. Each of these developments preceded major technological developments, including the stone age, the bronze age, and the industrial revolution.

Fast forward to modern times, and the agriculture industry faces a monumental problem; the destruction of the environment. Simply put, the overuse of fertilisers, herbicides, and pesticides is destroying wildlife and local habitats and significantly affecting ecosystems that enable life to flourish. As such, engineers are desperately trying to develop solutions that eliminate the need for farmers to use these chemicals.

One solution has been found in using automatic laser targeting of weeds. A robot fitted with a laser and imaging system can identify non-crop plants and then blast those plants with a laser. This kills the weed, preventing precious nutrients and water from being diverted from crops. While these systems are still in development, they have proven to be effective, demonstrating how robotics will be a key technology for the future of farming. 

But it’s not just weeds that need removal. Even some intentionally planted crops are required to be eliminated as too many crops in a given area will try to compete for resources, and therefore result in stunted growth (there is a very interesting story about a soviet scientist who caused massive starvation by instructing farmers to tightly grow crops thinking that plants are socialists, and thus won’t compete with each other).  

This idea of eliminating competing crops is called thinning, which is a crucial step in cultivating some crops. But trying to thin crops using current technologies is challenging as it can be hard for a machine to decide which crops should be eliminated. For example, smaller plants may be considered weak, but larger plants could suffer from discolouration in their leaves, indicating that they may be suffering from a disease. 

Carbon Robotics develops thinning laser technology

Recognising the challenges of crop thinning in automation, Carbon Robotics announced the development of crop thinning technology that works with their pre-existing LaserWeeder system. According to Carbon Robotics, their technology can be used to successfully identify suitable crops for thinning, including lettuce, spinach, and broccoli. 

The ability to thin plants is provided as a software update to the wedding fleet, meaning that operators of the robotic systems do not need to make any hardware adjustments or purchase new equipment. Additionally, the new update also allows for operation during night and day, eliminating farms from being dependent on sunlight. This not only helps to reduce labour costs with weeding and thinning but also helps to massively improve efficiency and crop yield. 



It has been claimed that the LaserWeeder system enables farmers to break even on costs within 1 to 3 years of use, but more data would be needed to prove this claim. However, this claim isn’t unreasonable as farmers can eliminate their need for herbicides and significantly reduce water and fertiliser use.

Will robotic systems dominate the agriculture industry?

As resources become increasingly constrained, and the desire to protect the environment grows, robotic systems capable of identifying and eliminating weeds and pests will undoubtedly prove instrumental. The use of renewable energies located at farms (via solar and wind) could help provide energy to agricultural equipment, and the training of AI models in remote data centres can provide real-time updates to equipment, making robotic systems more capable of identifying better farming practices

Of course, robotic farm systems are still in their infancy, and most currently used equipment is still manually operated. There are systems that can operate independently, but they are often pulled around by a tractor or some other vehicle. But, developments in AI and robotics will undoubtedly begin to replace manual labour, improve yields, and lower costs, all of which will reduce the price of food worldwide. At that point, we will likely see another cultural and technological revolution when people have more time on their hands. 

Profile.jpg

By Robin Mitchell

Robin Mitchell is an electronic engineer who has been involved in electronics since the age of 13. After completing a BEng at the University of Warwick, Robin moved into the field of online content creation, developing articles, news pieces, and projects aimed at professionals and makers alike. Currently, Robin runs a small electronics business, MitchElectronics, which produces educational kits and resources.