This study evaluated the performance of the YOLOv12 object detection model, and compared against the performances YOLOv11 and YOLOv10 for apple detection in commercial orchards based on the model training completed entirely on synthetic images generated by Large Language Models (LLMs). The YOLOv12n configuration achieved the highest precision at 0.916, the highest recall at 0.969, and the highest mean Average Precision (mAP@50) at 0.978. In comparison, the YOLOv11 series was led by YOLO11x, which achieved the highest precision at 0.857, recall at 0.85, and mAP@50 at 0.91. For the YOLOv10 series, YOLOv10b and YOLOv10l both achieved the highest precision at 0.85, with YOLOv10n achieving the highest recall at 0.8 and mAP@50 at 0.89. These findings demonstrated that YOLOv12, when trained on realistic LLM-generated datasets surpassed its predecessors in key performance metrics. The technique also offered a cost-effective solution by reducing the need for extensive manual data collection in the agricultural field. In addition, this study compared the computational efficiency of all versions of YOLOv12, v11 and v10, where YOLOv11n reported the lowest inference time at 4.7 ms, compared to YOLOv12n's 5.6 ms and YOLOv10n's 5.9 ms. Although YOLOv12 is new and more accurate than YOLOv11, and YOLOv10, YOLO11n still stays the fastest YOLO model among YOLOv10, YOLOv11 and YOLOv12 series of models. (Index: YOLOv12, YOLOv11, YOLOv10, YOLOv13, YOLOv14, YOLOv15, YOLOE, YOLO Object detection)

Washington State University scientist Manoj Karkee recognized by Connected World magazine for his work with agriculture technology. Washington State University (WSU) scientist Manoj Karkee was recently named a 2019 Pioneer for his work in artificial intelligence and internet of things by Connected World magazine. Among his projects, Karkee is building apple-picking robots, smart irrigation systems for grapes and fruit trees, flying drones to deter birds from fruit crops, and machines to bundle red raspberries. Just as important as building machines, he develops artificial intelligence for field agriculture, creating the software that tells agricultural robots how to do their work. He was one of 11 scientists in the U.S. and Canada to receive the recognition from the online journal.

I don't think sci-fi saw this coming. For so long, futuristic books and films have promised us robots like C-3PO that translate alien languages and assist us in hijinks. Or ones like Rosie that clean our houses. Or, on the other end of the spectrum, robots that level our houses and destroy humanity. The reality of modern robotics couldn't be more different.

Dan Wheat/Capital Press Joe Davidson, a Washington State University mechanical engineering doctoral student, demonstrates use of a robotic apple picker at a WSU field day in Prosser, Wash., on Sept. 17. Such a device could be a big labor saver for the apple industry. A robot able to pick apples fast enough and gently enough to be economically viable could be a huge boost to the apple industry in labor savings and in meeting labor shortages. Manoj Karkee, associate professor of biological systems engineering at the Washington State University research station in Prosser, and a new company, Abundant Robotics, of Menlo Park, Calif., tested robotic picking in Central Washington orchards last year and again this fall. About a dozen companies around the world and another three to five research groups are working on the robotic harvest of apples, citrus, bell peppers, cucumbers and other fruits and vegetables predominantly picked by hand, Karkee said.