Control of Sorghum halepense in soybean using the ATAR smart sprayer system

The ATAR smart sprayer is an innovative system developed by the startup company ATAR AGTEK d.o.o. It enables precise identification and treatment of weeds and can be easily integrated with any type of sprayer. The system consists of cameras that record area in front of the sprayer as it moves. The images are transmitted to an advanced artificial intelligence chip, which analyzes them in real time and activates the necessary nozzles to target and effectively treat only the detected weeds. In 2023 and 2024, the precision of this system in recognising and controlling Sorghum halepense from rhizomes in soybean crops was tested in field trials. Fields with a highly uneven distribution of S. halepense plants, where control was considered justified were selected for testing. In both trials, a clethodim-based product (144 g A.I. ha-1) was applied either broadcast over the entire area of the plots or only targeted where the presence of this weed was detected. Herbicide was applied post-emergence at stages 13-31 of the BBCH scale for S. halepense and at stages 12-13 of the BBCH scale for soybean. In 2023, a trial was conducted on a large plot without replication, while the 2024 trial was set up according to a randomized block design with three replications. To confirm the effective recognition and treatment of weeds in each basic plot, water-sensitive papers placed horizontally next to the weed plants and the SnapCard application was used to analyze percentage coverage. The most important parameter was efficiency, defined as the proportion of controlled weed plants 15 days after herbicide application, evaluated on a 400 m2 area in the middle of each basic plot. For each basic plot, a map of the treated areas was created and the percentage of the treated area was calculated. Analysis of water-sensitive papers confirmed that weeds were fully identified and treated using this innovative approach. The treatment maps showed that in 2023 only 19% of the total area was treated, while in 2024 this ranged from 29-75%. These differences depended on the number and spatial distribution of weeds, as well as the setting of the required treatment area in front of and behind the detected weeds in order to achieve maximum system reliability. By analyzing the number of controlled and uncontrolled (newly sprouted) plants from the rhizome 15 days after herbicide application in the 2023 trial, the effectiveness of S. halepense control was 92.5%. In 2024, similarly high efficacy was achieved with all four tested treatments: at speeds of 6 and 8 km h-1 whole-surface treatments achieved 91.2% and 94.2% control, while spot spraying achieved 90.8% and 93.5% at the same speeds. The results confirm that this system can provide reliable control of S. halepense with significant herbicide savings, thereby reducing the risk to non-target organisms and the environment.