An AI Edge Computing-Based Robotic Arm Automated Guided Vehicle System for Harvesting Pitaya

Abstract

This paper proposes an intelligent agricultural robotic arm automated guided vehicle (AGV) system, which is based on AI edge computing for harvesting pitaya. The proposed system is used to detect the growth of pitaya fruit in the natural environment. The proposed system uses computer vision technology to recognize the position of the ripe pitaya fruit in real-time. The robotic arm can grasp the pitaya accurately. Furthermore, the YOLO v3-tiny model is adopted to analyze mature and unripe pitaya. Since the color of the immature fruit and the stem are similar, it produces high variability detection, which brings some challenges in this work.