Robotic grippers are becoming increasingly vital in modern agriculture, especially in tasks like harvesting delicate crops such as tomatoes, where precision and care are crucial. These advanced tools are designed to handle tomatoes without causing damage, significantly improving efficiency and reducing labor costs. Research on gripper robots for fruit picking continues to be developed using various methods in an effort to achieve accurate picking results. This study proposes a hybrid method that combines Finite State Machine (FSM) for behavior control with Fuzzy Logic Control (FLC) to optimize the positioning of the gripper. The system utilizes a PixyCam2 CMUcam5 for tomato detection, an Arduino microcontroller for image processing, and a servo mechanism to precisely align the gripper with the target. The experimental results confirm that each component functions as expected, with the gripper successfully performing actions such as idling, gripping, and placing in accordance with the FSM model. Furthermore, the FLC model was tested against simulations, resulting in error rates of 1.004% for the elbow angle and 0.826% for the base angle. The entire system was validated by comparing the performance of the system using FLC and non-FLC in ten tests, each with tomatoes placed in different positions. The results indicate that the proposed gripper, utilizing the FSM-FLC model, achieved a 100% success rate in grasping the target, significantly outperforming the FSM-non-FLC gripper, which achieved only a 20% success rate. These findings have important implications for the agricultural industry. The successful integration of the FSM and FLC models in robotic grippers paves the way for fully automated harvesting systems, potentially reducing costs and enhancing productivity.
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.