A Liquid Metal-Based Smart Insole with Stress-Concentrating Soft Protrusions for High-Fidelity Dynamic Motion...

Realizing high pressure sensitivity in intrinsically soft sensors is a critical challenge for precise human-machine interfaces (HMI), as achieving high sensitivity while maintaining the intrinsic flexibility and wearability of fully soft materials remains a significant challenge. Herein, we report a fully-soft, protrusion-integrated smart insole for comfortable, high-fidelity dynamic motion tracking. By utilizing protrusions as stress concentrators, the system overcomes silicone's incompressibility, inducing rapid geometric collapse of underlying liquid metal microchannels. Notably, parametric optimization of the protrusion geometry reduced the critical detection threshold (Fc) down to as low as 1 N. This significant reduction effectively minimizes the sensing dead zone, ensuring high sensitivity even in the low-pressure regime. Fabricated via a hybrid process of direct ink writing and monolithic molding, the sensor ensures a seamless interface and structural integrity without rigid inclusions. The wireless system collects stable signals at 60 Hz, and when integrated with a 1D-CNN, achieved a high classification accuracy of 96.62% in a mixed-subject scenario. Furthermore, the system demonstrated robust generalization with an accuracy of 79.4% in inter-participant tests, and superior personalized performance with over 99% accuracy in intra-participant tasks.