An evolutionary design method for magnetically-actuated origami soft robots

Abstract Magnetically-actuated origami soft robots (MAOSRs) have attracted substantial attention owing to their inherent merits, such as the ability for remote actuation, high flexibility, and fast response. However, the design of existing MAOSRs primarily relies on a trial-and-error approach, which is highly influenced by the expertise of researchers. The existing designs of MAOSRs mainly consist of conventional crease pattern and straightforward magnetization distributions, restricting the capacity of MAOSRs. To enable the programmed automatic design of MAOSRs that integrates the structure and actuation elements, we propose an evolutionary design framework in this work. The proposed method effectively tackles the optimal design of MAOSRs by concurrently considering the crease pattern, material mechanical properties, remnant magnetization distribution, and applied magnetic field. Two representative design problems, including shape-programming and maximizing target output under magnetic response, have been used to verify the applicability of the design method. Three sets of optimization cases and experiments, including a shape-programming origami arm, and the well-known chomper-based and square twist-based patterns, have been conducted to assess the effectiveness of the proposed method. Finally, pick-and-place and obstacle avoidance experiments were performed to evaluate the performance of the designed square-twist gripper.