Safflower Production Management ECOSYSTEM with AI harvester

Abstract. Safflower is a crop cultivated in the Middle East and is an annual plant of the Asteraceae family with a height of 0.5 to 1.0 m and a radius of 2.5 to 4.0 cm in early summer. Safflower is used for high-quality local foods cosmetics, dyes, but the harvesting is done by hand, therefore the work efficiency is low. In this study, we developed a cultivation management system to increase the efficiency of the process; from cultivation to harvest, aimed at revitalizing the production of safflower. Advancements in Artificial Intelligence and robotics has allowed for the exploration remote sensing methods that gather and analyses large amounts of data to guide precision management. A detecting unit for buds and flowers was developed using AI(Deep Learning) to count and analyze the location. This data is stored inGIS applicaiton to manage the growth of buds and flowers. The flower location data is used by the AI harvesting robot and robot arm controller to guide the harvesting unit to the flower. The harvesting unit cut the petals by rotary cutter, then vacuum the petals into a storage tank. Further work is aimed at improving the harvesting performance and going to enable the new feature of autonomous driving and harvesting in field. And it can be used as AI harvester has multi-feature robot as ECOSYTEM for variety of plants and crops, fruits precision management. We hope to help the elderly farmers to keep the production of safflower in Yamagata, Japan.