Inverse Co-Design of Mechanical And Sensory Properties in Soft Lattice Foams for Multifunctional Wearables.

Lattice structures exhibit a variety of topological geometries, which endow them with diverse mechanical properties and high design flexibility, and enable them to be integrated with sensing mechanisms to collect environmental or intrinsic information. However, the traditionally empirical and separated design leads to difficulties in the integration of multiple functions and limits the exploration of the extensive properties of lattice foams. In this work, a design methodology is proposed for the inverse co-design of the mechanical and sensory properties of lattice foams by training an inverse-design neural network and optimizing the layer laminating permutations. Physics-informed numerical models and hybridization principles are employed to generate training data for the inverse mechanical design of hybrid foams. Subsequently, the mechano-electrical relationships are defined and utilized to predict and optimize the sensory behaviors of the hybrid foams. As a proof-of-concept example, a smart knee pad is developed to meet a specified target protective function and optimized sensory behavior, demonstrating not only its inverse-designed mechanical response but also its sensing capability from light interaction forces with users to heavy impact forces from collisions. This concept allows for more data-driven analytical approaches to satisfy the various needs for multi-functional wearable devices.