The development of iontronic skin (I-skin) capable of ultrafast sensing in a wide pressure range, comparable to human skin, is of paramount importance for intelligent robotics. However, this remains a major challenge due to the lack of iontronic array architectures that can achieve ultrafast readout and crosstalk-free under large capacitance response generated within a wide pressure range. Here, we report a frequency-coding architecture of artificial ion mechanoreceptor skin (AIM-skin) that can provide a universal mode of iontronic array sensing and bypass the dependence of complex integrated back-end interface electronics. Notably, the successful implementation of orthogonal frequency coding in the AIM-skin with high sensitivity and ultrawide pressure range achieve ultrafast parallel readout for the spatiotemporal mechanical stimuli. Furthermore, the parallel zero-potential mechanism (PZPM) of the architecture effectively mitigates electrical crosstalk between sensing units. We have demonstrated that combhination of proposed device and deep learning has a broad application prospect in intelligent human-machine interaction and real-time dynamic robotic manipulation.

开发能够在较宽压力范围内超快传感（与人类皮肤相当）的离子电子皮肤（I-skin）对于智能机器人技术至关重要。然而，由于缺乏能够在宽压力范围内产生的大电容响应下实现超快读出和无串扰的离子电子阵列架构，这仍然是一个重大挑战。在这里，我们报告了一种人工离子机械感受器皮肤（AIM-skin）的频率编码架构，它可以提供离子电子阵列传感的通用模式，并绕过复杂的集成后端接口电子器件的依赖性。值得注意的是，在具有高灵敏度和超宽压力范围的 AIM 皮肤中成功实现正交频率编码，实现了时空机械刺激的超快并行读出。此外，该架构的并行零电位机制（PZPM）有效减轻了传感单元之间的电串扰。我们已经证明，所提出的设备和深度学习的结合在智能人机交互和实时动态机器人操纵方面具有广阔的应用前景。