To simplify the process of weeding, researchers recommend genetically modifying crops to have distinct colors or unique leaf shapes. This would make it easier to differentiate them from wild and weedy plants. One approach is to modify the crops’ DNA to produce pigments commonly found in other plants, such as anthocyanins (which give blueberries their color) or carotenoids (which give carrots their orange hue). Once this is achieved, weeding robots could be programmed to use machine learning to selectively remove the unwanted weeds.
To enhance the identification of weeds, the researchers recommend using gene editing methods to modify the crops’ characteristics, such as their leaves, to make them more distinguishable from wild and weedy plants. This may involve modifying the crops’ genetic makeup so that they produce pigments commonly found in plants, such as anthocyanins for a blue color or carotenoids for an orange hue, making it easier to differentiate them. They also suggest the use of weeding robots that can be trained to selectively remove only the unwanted plants through machine learning. The scientists detail their proposed approach in the April 17 issue of the journal Trends in Plant Science.
“The ultimate goal is to cultivate a new variety of crops that are environmentally sustainable, high-yielding, and supportive of eco-friendly agricultural practices,” the researchers state.
However,r, De novo crops that have been domesticated from wild plants are likely to closely resemble their wild counterparts, which would pose difficulties for weeding. One possible solution would be to incorporate herbicide resistance genes into the de novo crop and then use herbicides to eliminate the weeds. However, the researchers propose an alternative approach that involves combining genetic engineering with AI technologies to develop visually distinct de novo crop plants that can be easily distinguished from weeds by robotic weeders.
“Distinguishing these new crops from their less productive and closely related wild plants could present tremendous challenges for weed control,” the researchers write. “Utilizing AI technologies to visually differentiate between the crops and weeds could be a valuable strategy.”Using gene editing to improve the ability of robots to identify and remove weeds could effectively solve this problem.”
Instead of introducing genes from other organisms into the new crops, the researchers suggest modifying the plants’ genetic makeup so that they produce pigments that are already present in many plants. For example, they propose including anthocyanins, which create red, purple, and blue colors in plants like berries and purple cabbage, and carotenoids, which are responsible for the yellow, orange, and red colors in various plant parts such as carrots, peppers, and leaves, where they play a crucial role in photosynthesis.
The researchers state that manipulating these crucial genes would greatly improve the ability to differentiate between newly domesticated crops and their wild counterparts. In addition to aiding in visual differentiation, these pigments could also provide other advantages for both plant and human health. Plants with anthocyanin accumulation are known to have increased resistance to herbivory, fungal diseases, bacterial infections, heavy metal toxicity, and other environmental stresses, while carotenoids serve as a source of provitamin A in the human diet.
Because of these beneficial traits, plants rich in anthocyanin offer numerous advantages.The researchers suggest that using leaf shapes as a way to differentiate between domesticated crops and weeds could have practical applications in agriculture. Another option could be to modify the leaf structure of newly developed crops to make them simpler or more complex than their wild counterparts, such as by introducing mutations in the genes responsible for adding or removing leaf lobes. Additionally, new crops could be engineered to have distinct seed colors or shapes to facilitate sorting after harvest. Further research is necessary to determine whether these modifications would affect the vitality of the crops.This involves testing whether these pigments disrupt the process of photosynthesis or the ability of plants to withstand stress. There is also a need for further research to enhance remote sensing methods and determine the most effective ways to train automated weeding systems to identify the new characteristics of newly cultivated crops.
Journal Reference: