Non-destructive fruit firmness evaluation using a soft gripper and vision-based tactile sensing

• DOI: 10.1016/j.compag.2023.108256
• Publication Date: 2023-09-30T09:30:39Z
• Authors: Jiahao Lin; Qing Hu; Jinming Xia; Liang Zhao; Xuan Du; Shanjun Li; Yaohui Chen; Xing Wang
• Keywords: Soft Robotics; Tactile sensor

As fruit firmness is a crucial characteristic associated with the maturity level, its accurate estimation of great importance to post-harvest processing and wholesale in industry. Benefiting from advances soft robotics, gripper simultaneous compliant deformation tactile sensing proposed this study for classification. The design inspired by fin ray effect can achieve active deformation, which helps simplify actuation system improve delicate manipulation capability. Finite element modelling, along experimental tests, first utilized validate gripper's feasibility safe grasping, respiratory tests are then conducted further demonstrate non-destructive nature. Moreover, fruit–gripper interaction captured visual sensors processed using an attention-based CNN–LSTM algorithm predict information. Tomatoes nectarines chosen as sample validation. R2 values their prediction 0.795 0.753, accuracy grading 84.6% 81.5%, respectively. In general, provides promising solution both grasping evaluation, it expected be integrated into automated production lines pack based on different levels.