Towards Bespoke Soft Grippers through Voxel-Scale Metamaterial Topology Optimisation

Soft grasping has become the accepted standard to grasp geometrically complex and deformable objects; however, the in the absence of efficient design tools, ‘universal’ designs have proliferated. Bespoke design approaches, which exploit morphological computation to increase contact area and grasp success and reduce contact force, have been limited by the complexity of modelling the grasping behaviour of geometrically uncertain deformable designs. We present a framework for bespoke soft gripper design based on functionally graded topology optimisation, which exploits both macroscale geometry and microscale material properties to generate high-quality grasps. Using a model of the grasped object, we simulate the interactions between the deformable gripper and object to find the optimal material properties and material distribution across a voxelised design domain. We develop a material interpolation method which generates custom metamaterials by combining hard and soft materials at high resolution and explore multi-material polyjet 3D printing to produce functionally graded soft grippers with the metamaterials. We show that our bespoke design method produces unique geometries tailored to specific objects, and that the combination of bespoke geometric design and localised metameterials produces unique designs which enhance grasp performance.