Programmable auxetic architectures in soft robotics: A three-dimensional review on design, additive manufacturing, and optimization

In this comprehensive review, literature on programmable auxetic architectures (PAAs) in soft robotics is synthesized, highlighting progress in design, additive manufacturing (AM), and optimization of these structures. Real-time mechanical property modulation in Poisson’s ratio, deformation modes, and load bearing capacity can be performed by design parameterization, multiscale hierarchy, utilization of smart materials like shape-memory polymers, and stimuli-responsive mechanisms. Design policies include unit-cell engineering for specific auxetic effects, topology optimization, genetic algorithms, finite element analysis (FEA), machine learning for inverse design, and multi-objective frameworks that equalize stiffness, durability, and actuation efficiency. Mechanical testing, digital image correlation, FEA validation, and digital twins are performance characterization techniques that have direct application in soft grippers for adaptive manipulation, deployable structures, wearable exosuits, locomotion for terrain traversal, and biomedical devices. Further, materials such as soft polymers, carbon nanotubes, composites, and hybrids can be used to improve functionality. However, there are gaps in dynamic or multi-axial testing, standardization, biocompatibility, and scalability. Future research should focus on artificial intelligence (AI)-driven co-design, sensor-actuator fusion, multi-physics simulations, and standardized protocols to improve the technology readiness level (TRL) for industrial soft robotic applications.