Additive manufacturing provides a rapid manufacturing method for a variety of materials with different applications. Thermoplastic Polyurethane (TPU) is a soft polymer material that can be 3D printed. In this work, we explore the mechanical properties of a 3D printed grid pattern structure with TPU. By changing the pattern’s cell size and wall thickness parameters, we control the density of the grid lattice and, as a result, the bulk elastic modulus of the structure. We compare simulation and physical compression tests and conclude that the bulk elastic modulus of a print is related to the infill percentage according to a cubic relationship, with higher infill percentage samples resulting in higher elastic moduli. The precise cell size and wall thickness parameter values are minor influences comparatively. The elastic moduli of the resulting samples span from 0.36 MPa with 23.44% infill to 21.83 MPa with 75% infill, compared to an elastic modulus of 64.31 MPa when printing at 100% density. We also explore other factors such as the sample size, the printer, the build orientation, and the sample geometry. The results have uses in a variety of applications, including a custom linear spring, a bistable gripper, or a soft robot finger.

Grasping is an essential task in robotic applications and is an open challenge due to the complexity and uncertainty of contact interactions. In order to achieve robust grasping, systems typically rely on precise actuators and reliable sensing in order to control the contact state. We propose an alternative design paradigm that leverages contact and a compliant bistable mechanism in order to achieve "sensing" and "actuation" purely mechanically. To grasp an object, the manipulator holding our end effector presses the bistable mechanism into the object until snap-through causes the gripper to enclose it. To release the object, the tips of the gripper are pushed against the ground, until rotation of the linkages causes snap-through in the other direction. This push-on pushoff scheme reduces the complexity of the grasping task by allowing the manipulator to automatically achieve the correct grasping behavior as long as it can get the end effector to the correct location and apply sufficient force. We present our dynamic model for the bistable gripping mechanism, propose an optimized result, and demonstrate the functionality of the concept on a fabricated prototype. We discuss our stiffness tuning strategy for the 3D printed springs, and verify the snapthrough behavior of the system using compression tests on an MTS machine.