Hand-Picking Dynamic Analysis for Undersensed Robotic Apple Harvesting

Abstract

<abstract> This article evaluates hand-picking methods as candidate grasping techniques for implementation in a robotic system designed to harvest apples. The standard method of hand-picking apples is highly selective to the apple‘s orientation and stem location. However, sensory detection of the fruit‘s orientation and stem while the apple is on the tree is a challenging problem requiring significant computation time. In this study, four picking techniques that do not require knowledge of fruit orientation were applied to five apple varieties growing in several different cultivation systems. The sensors used during hand-picking included force sensors and an inertial measurement unit. Experimental results were obtained for normal contact forces during a three-fingered power grasp as well as the angle of rotation around the axis of the forearm. Field data and controlled laboratory experiments show that fruit separation can be clearly detected. Accelerometer measurements were also used to calculate the average distance to fruit separation, which varied from 3 to 7 cm. The optimum picking method relative to stem attachment was identified for each apple variety.