High-Tolerance Soft-Rigid Gripper for Low-Damage Robotic Strawberry Harvesting

Abstract

This letter presents the design, modeling, and experimental validation of a novel soft-rigid hybrid gripper for robotic strawberry harvesting that addresses key challenges in damage prevention and obstacle handling. The design incorporated circumferentially arranged pneumatic chambers integrated with rigid supports, enabling morphology-adaptive grasping with high localization tolerance and reduced environmental interference. An integrated airflow module was developed to clear surrounding obstacles prior to grasping. A Support Vector Machine (SVM)-based classifier was proposed for state recognition, aiming to close the loop for detachment. Field experiments under manual operation demonstrated that the gripper achieved a 93.4% success rate, with only 2.9% damage to fresh-market strawberries. Also, the gripper maintained a 93% swallowing success under translational and angular errors of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula>18 mm and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula>27°, respectively. For clustered strawberries, the obstacle-clearing function increased the autonomous harvesting success rate from 85.8% to 89.2%. Comparative studies verified that the gripper required only 18.6 kPa peak pressure to detach fruit, achieving damage control comparable to or better than manual harvesting. This work offers a robust solution for delicate fruit harvesting in clusters.