Advancing Robotic Fruit Picking: Adaptive End-Effector Approach

In modern agriculture, the demand for efficient fruit picking methods is constantly increasing due to labor shortages and the need to improve productivity. Traditional fruit picking methods often rely on manual labor, which is labor-intensive, time-consuming, and prone to inconsistency. This study aims to develop a novel end effector for agricultural fruit picking robots, which improves the efficiency and accuracy of fruit picking through highly flexible and adaptive design. It can effectively handle fruits of diverse types, shapes, and hardness. By adopting a three-finger scheme and independent servo motor control, the simulated movements of human hands are realized. Shape memory alloy (SMA) and electroactive polymer (EAP) as intelligent driving materials, as well as flexible materials such as silicone, are explored to optimize the grip performance and fruit protection of the actuator. High-precision visual and tactile sensors are also integrated to support the accurate identification of fruits of different maturity. The performance of the actuator is evaluated through advanced data analysis techniques, including descriptive statistics, hypothesis testing and regression analysis. The findings suggest that the proposed design optimization measures based on data analysis results can further enhance the adaptability and efficiency of the robot in future applications, thereby addressing the challenges faced in modern agricultural practices and contributing to increased productivity and sustainability in the agriculture sector.