Intelligent Self-Powered Gait Monitoring Using Triboelectric Nanogenerators: From Device Design to AI-Driven Health Prediction

Abstract The rapid progression in smart wearable healthcare technologies has intensified the demand for sustainable, maintenance-free, and intelligent sensing systems for continuous human motion monitoring. Among the various sensing systems, triboelectric nanogenerators (TENGs) emerge as a promising solution for self-powered gait monitoring due to their ability to efficiently harvest biomechanical energy from human body motion. This review provides a comprehensive overview of intelligent self-powered gait monitoring systems based on TENGs, spanning from device design to data-driven health prediction. We discuss various device designs for the analysis of gait monitoring and motion sensing and fabrication strategies that enhance sensitivity, flexibility, and durability for wearable applications. Subsequently, we examine the integration of TENG-based sensors into smart footwear and wearable platforms for real-time gait analysis, highlighting key performance metrics and signal characteristics. The review further explores the convergence of TENG sensing with artificial intelligence (AI) techniques, including machine learning and deep learning algorithms, for accurate gait pattern recognition, anomaly detection, and early diagnosis of osteoarthritis, flat foot, and Parkinson’s disorders. Finally, current challenges such as signal variability, system integration, data reliability, and scalability are critically analyzed, along with future perspectives toward personalized, AI-driven healthcare systems. This work aims to provide a unified framework for the development of next-generation intelligent, self-powered gait monitoring technologies and paves the way toward artificial intelligence of things (AIoT).