AI-Assisted Automated Identification of Human Activities Using Ficus religiosa Leaf-Based Triboelectric Nanogenerator

Movement monitoring and effective identification of different actions are the keys that help in fitness services, health status, clinical studies, etc. In this technological era, Internet of Things (IoT) technologies, including smart wireless devices and sensors, are very effectively used for monitoring human activities, but the demand for sustainable and green power sources is a crucial issue with these devices. Triboelectric nanogenerators (TENG) are proven to be promising applications in these devices because they harvest energy from the surrounding environment and eliminate the use of batteries as power sources. As a green energy source, this study emphasizes the fabrication of biodegradable materials-based TENGs, which are eco-friendly and are related to clean and green energy as per the UN's sustainable development goals SDG 7 (affordable and clean energy). In the present work, a natural Ficus religiosa leaf (FRL) of the F. religiosa tree is used in designing and fabricating a TENG (FRL-TENG). Also, an approach is discussed to compare the performance of FRL-TENG with TENGs fabricated from other waste biodegradable materials such as garlic tunic, onion tunic, and eggshell membrane (ESM). During the experimental study, it is observed that the FRL-based TENG produced maximum voltage in comparison to other material combinations selected in this study. The generated electric output from these TENG combinations is also used to power an array of tens of green-light-emitting diodes (LEDs). Furthermore, this paper also proposes the use of FRL-TENG as a wearable sensor to collect information and monitor the physical activities of the user, viz., walk, jump, and run. To recognize the movement status, the FRL-TENG sensor is integrated with an extra randomized tree-based machine learning model for accurately distinguishing the user's three activities with an accuracy of 96%. The work showcases an innovative approach to encourage customized uses of TENG sensors in human motion monitoring and permits the development of intelligent, self-powered systems for new applications.