Computational design of passive grippers

This work proposes a novel generative design tool for passive grippers -- robot end effectors that have no additional actuation and instead leverage the existing degrees of freedom in robotic arm to perform grasping tasks. Passive are used because they offer interesting trade-offs between cost capabilities. However, designs limited types shapes can be grasped. use rapid-manufacturing optimization expand space passively Our algorithm takes an object its positioning with respect generates 3D printable gripper stably pick up. To achieve this, we address key challenge jointly optimizing shape insert trajectory ensure stable grasp. We evaluate our method on testing suite 22 objects (23 experiments), all which were evaluated physical experiments bridge virtual-to-real gap. Code data at https://homes.cs.washington.edu/~milink/passive-gripper/