Stretchable Polymorphic Electrochromic Textile Electronics for Deep Learning-Assisted Self-Adaptive Camouflage

Electrochromic (EC) textile electronics, characterized by their seamless system integration, advanced intelligence, and intrinsic wearability, are promising to become the second skin of human beings for conformal real-time display, imaging, sensing, communicating, and human-machine interaction. However, it remains a huge challenge to simultaneously achieve EC textile electronics with admirable color-changing performance, reliable mechanical properties, remarkable environmental stability, and precise hierarchical control. Herein, a stretchable all-solid-state polymorphic EC textile electronics system, featuring fast response (≈3 s), high color contrast (≈58.1%), long-term operation (>12 000 s), and excellent coloration stability under large deformation and freezing is developed based on liquid-free poly(urethane-urea) elastomeric ionic conductors and is used to implement deep learning-assisted self-adaptive camouflage on human bodies in response to dynamic environments. This excellent performance is derived from the elastomeric ionic conductors designed with a supramolecular-covalent dual cross-linking network, which possess high ionic conductivity, superb stretchability, tissue-level softness, robust adhesion, self-healing property, and wide temperature adaptability. Moreover, a lightweight deep learning model, MobileNetV2 is developed to process environmental information, establish the coloration strategy and trigger mechanism for camouflage assignments. This work presents a biomimetic self-adaptive camouflage system using a smart glove as an example implementation, promoting a paradigm shift in next-generation wearable electronics.