AI-driven design of powder-based nanomaterials for smart textiles: from data intelligence to system integration
Zihui Liang, Yun Deng, Zhicheng Shi, Xiaohong Liao, Huiyi Zong, Lizhi Ren, X. L. Li, Xinyao Zeng, Peiying Hu, Kexiang Wei, Bing Wu, Kai Wang, Jin Qian, Weilin Xu, Fengxiang Chen
Abstract
Artificial intelligence (AI) is emerging as a transformative enabler in the development of smart textile systems, particularly those integrating powder-based functional materials. This review highlights recent progress in AI-guided design of carbon nanomaterials, metallic nanoparticles, and framework-based powders for applications in energy harvesting, intelligent sensing, and robotic actuation. Machine learning techniques, including supervised learning, transfer learning, and Bayesian optimization are discussed for accelerating materials discovery, enhancing integration strategies, and enabling real-time adaptive control. Emphasis is placed on how AI enables multifunctional, wearable platforms that sense, process, and respond to environmental and physiological cues with high accuracy and autonomy. Representative breakthroughs in soft robotics, haptic interfaces, and assistive devices are presented, demonstrating the synergy of AI and responsive textiles. Finally, the review outlines key challenges related to data scarcity, model generalizability, manufacturing scalability, and sustainability, while proposing future directions involving multimodal learning, autonomous experimentation, and ethics-aware design. This work offers a comprehensive outlook on next-generation AI-driven textile systems that seamlessly integrate intelligence, functionality, and wearability.