Deep-learning-assisted printed liquid metal sensory system for wearable applications and boxing training

Liquid metal (LM) is a promising material for high-performance soft electronics due to its high electrical conductivity and deformability. However, rapid patterning LM to achieve a highly sensitive sensory system remains challenging. This paper reports a rheological modification strategy of LM and strain redistribution mechanics to improve the sensitivity and printability of LM sensors. By incorporating SiO₂ particles, the LM-SiO₂ composite's modulus, yield stress, and viscosity are enhanced, enabling 3D printability. The printed LM-SiO₂ composite sensors exhibit excellent mechanical flexibility, robustness, and sensing performance. Integrated onto a tactile glove, these sensors, with deep-learning assistance, achieve 90.5% accuracy in recognizing boxing punches (jab, swing, uppercut, and combinations). This system has potential applications in smart sports training, soft robotics, and human-machine interfaces.

Ye Qiu, Zhihui Zou, Zhanan Zou, Nikolas Kurnia Setiawan, Karan Vivek Dikshit, Gregory Whiting, Fan Yang, Wenan Zhang, Jiutian Lu, Bingqing Zhong, Huaping Wu, Jianliang Xiao