In the past decade, a rich repertoire of soft robots, designed from biomimetic and intuitive approaches, has been developed to overcome challenges faced by their rigid-bodied counterparts. However, these design approaches are greatly limited by the designers' experience and inspiration. In this article, the structural design problem is mathematically modeled under the framework of topology optimization, and solved by a new implementation tool that combines Abaqus/CAE and Matlab coding. Herein, a pneumatic soft gripper with two identical fingers was developed as a practical application. To fulfill the grasping task, each gripper finger is optimized to achieve its maximal bending deformation. The optimized gripper fingers are in high consistence with human fingers as indicated by pseudo-joints. Thereafter, the optimized gripper fingers are directly fabricated by three-dimensional printing technique with unprecedented fidelity regardless of high geometric complexity. Experimental results show that the gripper can grasp an elastic balloon, and each gripper finger is able to undergo a [Formula: see text] free travel bending and exert 0.23 N grasping force upon 0.06 MPa actuation pressure. The proposed approach is freely extendable to develop other types of soft robots and this represents an important step toward the goal of designing and fabricating soft robots automatically.
Keywords: 3D printing; soft gripper; topology optimization.