Stiffening iron particles to modulate physical interactions.

Variable rigidity is vital for autonomous hardware systems to interact with unstructured environments, which allows them to be both soft and rigid depending on their motor intent. Extensive research has focused on granular jamming, owing to its excellent shape adaptability for irregular three-dimensional objects. However, previous approaches relying on pneumatic actuation suffered from limitations, such as slow jamming transitions and heavily tethered system designs. Here, we propose a magnetic field-based granular jamming principle, aiming to achieve rapid and precise stiffness tunability with an electrically driven modular hardware design. Our proposed concept incorporates a magnetorheological membrane, size-mixed soft magnetic particles, an optimized electromagnet, and a thin-film force sensor. Owing to the inherent nature of electromagnetic systems, the experimental results demonstrate a rapid response within 0.1 sec and precise stiffness tunability. Furthermore, we incorporate this technology into robotic applications, including grasping, locomotion, and tangible user interfaces, to demonstrate the enhanced performances of practical systems even with additional functions like in-fingerpad manipulation. The magnetic jamming concept and its design principle could be extended to other jamming geometries. We anticipate that this technology has promising applications in robotics, effectively modulating system-environment physical interactions.