Full Body-Worn Textile-Integrated Nanomaterials and Soft Electronics for Real-Time Continuous Motion Recognition Using Cloud Computing

Recognizing human body motions opens possibilities for real-time observation of users' daily activities, revolutionizing continuous human healthcare and rehabilitation. While some wearable sensors show their capabilities in detecting movements, no prior work could detect full-body motions with wireless devices. Here, we introduce a soft electronic textile-integrated system, including nanomaterials and flexible sensors, which enables real-time detection of various full-body movements using the combination of a wireless sensor suit and deep-learning-based cloud computing. This system includes an array of a nanomembrane, laser-induced graphene strain sensors, and flexible electronics integrated with textiles for wireless detection of different body motions and workouts. With multiple human subjects, we demonstrate the system's performance in real-time prediction of eight different activities, including resting, walking, running, squatting, walking upstairs, walking downstairs, push-ups, and jump roping, with an accuracy of 95.3%. The class of technologies, integrated as full body-worn textile electronics and interactive pairing with smartwatches and portable devices, can be used in real-world applications such as ambulatory health monitoring via conjunction with smartwatches and feedback-enabled customized rehabilitation workouts.