Design and optimization of magneto-active elastomer (MAE) unimorphs using genetic algorithms

Magneto-active elastomers (MAEs), a type of smart material, deform in response to external magnetic fields owing to their flexible, untethered, and reconfigurable capabilities. These capabilities are useful for biomedical devices, robotic systems, and actuators for which control over shape morphing, extent of deformation, and force output during actuation are desirable. However, programming the MAE to realize targeted functionalities such as complex deformations (e.g., biomimetic motions), or manipulating force (e.g., grippers) is challenging due to inherent trade-offs between deformation and force generation, where greater flexibility enables larger displacements but typically reduces force, necessitating a formal design and optimization approach. This work introduces a systematic design and optimization approach to program MAEs to achieve complex shape morphing and predictable force-displacement responses. The proposed approach enables optimization for target objectives by integrating previously developed analytical models for predicting the MAE actuation performance into evolutionary algorithms, including genetic algorithms and the multi-objective non-dominated sorting genetic algorithm II (NSGA-II). The optimization output includes the geometry and magnetic properties of the MAE to achieve target deformations or forces as closely as possible when actuated by contactless magnetic fields. Specifically, the approach is used to minimize deviations between the actuated MAE and a predefined target shape and to examine the performance trade-offs between free deflection and blocked force. This research offers a promising avenue for revolutionizing traditional static structures by incorporating smart materials capable of altering their shapes with customized and predictable displacement and actuation force profiles.