Conventional industrial grippers that grip a flat object generally hold objects by using suction or electrostatic force. However, these grippers have limitations when gripping thin, flat, and flexible objects, such as films and flexible printed circuit boards (FPCBs), due to their undefined shape and high flexibility. This paper proposes a soft gripper that can grip flexible and thin objects by utilizing directional adhesives and a compliant mechanism. The directional adhesive pad is fabricated by a three-dimensional (3D) printing process for cost-effective and environment-friendly manufacturing. However, fabrication by 3D printing has disadvantages in terms of the quality of the adhesive surface. An additional coating process presented in this study compensates for the low resolution of 3D printing by improving smoothness. Moreover, an additional coating process is a simple approach for developing directional adhesives with enhanced adhesion strength by deforming the tip shape without a sophisticated fabrication process. The adhesion of adhesives with curved pillars is enhanced compared to adhesives with simple wedge-shaped pillars. The maximum normal adhesion force of the gripper is measured to be 0.47 N (1.57 kPa), and 95% of the initial adhesion is retained after ten thousand attachment/detachment cycles. The adhesion force can be recovered by the cleaning process when the contaminant is attached to the adhesive. The final demonstration shows that the gripper can handle various objects for potential applications such as in green-environmental industries.
This work was supported by the Robotics Core Technology Development Project (20000512) funded by the Ministry of Trade, Industry and Energy (MoTIE, Korea), and the National Research Foundation of Korea (NRF) grant funded by the Korean Government (2020M1A3B8084924, 2019R1F1A1063066).This work was supported by the Robotics Core Technology Development Project (20000512) funded by the Ministry of Trade, Industry and Energy (MoTIE, Korea), and the National Research Foundation of Korea (NRF) grant funded by the Korean Government (2020M1A3B8084924, 2019R1F1A1063066). The authors would like to thank Mr. kyungmin baek, an undergraduate researcher of the MOST Lab, for his contribution to robot arm manipulation.