Development of a Button Mushroom Harvesting Robot using Expert System and Image Processing in Shelf Cultivation Method

Among the agricultural products, mushroom is one of the best candidates for robotic harvesting methods. One of the main problems of mushroom growers is a critical need for labor within the specified time for harvesting mushrooms, so mushroom growing is labor intensive. In this research, we attempted to develop and apply a harvesting robot for this crop to reduce the problems that growers face in terms of harvesting labor. For ease of end-effector movement between shelves, the robot was developed by Cartesian Mechanism. This robot used image processing and a computer expert system to detect the position of mushrooms on the substrate on the shelf. The end-effector acts by using a suction cup and a non-shear mechanism to harvest mushrooms from the substrate. By testing this robot on the substrate, we could harvest perfect whole mushrooms on average of ۸۱.۵% of mushrooms on the prepared substrates. Harvesting time per mushroom was obtained ۱۲.۴۵s, which is the amount of respective time of ۸.۴۵s more than harvesting time by labor hands, but in robotic harvesting, robots unlike humans can work ۲۴ hours a day, continuously during growing time. An expert system also could be a valuable asset to change the grower’s strategy in terms of harvested mushrooms size and quality based on customer needs in the market.

کلیدواژه ها:

Mushroom harvesting ، Robot ، image processing ، Expert System ، Shelf Cultivation Method