Design and Development of Prototype of Automated Strawberry Harvesting Mobile Robot Using MVS Integrated with Dynamic Vision CNN Model

The agricultural industry is increasingly adopting automation and technology to improve efficiency and yield. This paper presents a system design and development for an automated strawberry harvesting mobile robot using a machine vision system integrated with a dynamic Convolutional Neural Network (CNN) model. The system consists of several components, including a mobile robot equipped with a machine vision system, a dynamic CNN model for fruit detection and classification, and a robotic arm for fruit harvesting. The machine vision system is used to capture images of the strawberry plants, which are then processed by the dynamic CNN model to identify and classify the strawberries. The robotic arm is then controlled to harvest the ripe strawberries based on the classification results. The dynamic CNN model is designed to adapt to changes in the environment, such as variations in lighting conditions and the presence of obstacles. The model is trained on a large dataset of strawberry images, enabling it to accurately detect and classify strawberries in real-world conditions. The use of a dynamic CNN model ensures high accuracy and robustness in fruit detection and classification. The mobile robot is designed to navigate autonomously through the strawberry fields, using sensors and mapping algorithms to avoid obstacles and navigate to the next row of plants. The robotic arm is equipped with a gripping mechanism and a cutting tool, allowing it to gently harvest the strawberries without causing damage. In summary, the proposed system for automated strawberry harvesting using a machine vision system and dynamic CNN model offers several advantages, including increased efficiency, reduced labor costs, and improved fruit quality. The system is also scalable and adaptable to other fruit harvesting applications, making it a promising solution for the future of automated agriculture.