Topology Optimization of a Magnetic Soft Actuator for Enhanced Actuation Performance

Abstract:
Magnetic soft actuators, composed of compliant polymer matrices embedded with magnetic particles and actuated by external magnetic fields, offer remote, contactless operation and intrinsic compliance, making them ideal for biomedical and robotic applications. However, their performance is often limited by weak actuation forces and pronounced nonlinear behavior. This study investigates two distinct topology optimization strategies to overcome these limitations and improve the overall performance of a magnetic soft actuator. The first employs the adjoint method and is implemented in FEniCS platform to enable efficient gradient-based optimization. The second uses a simulation-driven approach based on ANSYS Maxwell and PyAnsys, enabling high-resolution field modeling and geometry-based design updates. Both methods are applied to two design objectives: maximizing magnetic force at a target area and achieving a uniform force profile over a vertical range. The optimized designs achieve a 200 % increase in actuating force or maintain a uniform force distribution with a deviation of ±20 % over a 6 mm range. Simscape-based dynamic modeling shows up to twofold improvements in both displacement and force output, while also enabling stable PI control. These findings demonstrate that magnetic topology optimization significantly improves both field quality and system-level performance, supporting advanced magnetic soft actuator design.