A Triboelectric Smart Carpet with an Optimized Braided Structure for Cruise Ship Monitoring Enabled by Deep Learning

In pursuit of comprehensive monitoring of personnel within cruise ship cabins, this study introduces a novel triboelectric smart carpet featuring an intricate braided design. This innovative carpet is allowed to gather individual information without the privacy concerns associated with video recording. The discourse begins with a detailed examination of the output characteristics inherent to the sensing unit, pixel, and the carpet as a whole. The findings suggest that the smart carpet not only precisely identify any walking trajectory but also enhances sensor scalability, showcasing its considerable promise for monitoring over extended floor areas. Further, this study integrates the smart carpet with Deep Learning (DL) techniques to validate its real-time monitoring capabilities, which encompasses state detection, personnel identification, and trajectory tracking. The outcomes demonstrate a commendably high level of precision, indicating a robust performance. Additionally, the smart carpet's design facilitates its integration with a suite of other sensors, positioning it as an instrumental asset for reinforcing maritime safety and fostering the development of Internet of Things (IoT) applications for maritime environments.