WEARABLE BIO-MONITORING FOR RESOURCE MANAGEMENT IN LUNAR EVA

Abstract

Lunar extravehicular activities (EVAs) are long missions in which astronauts perform various tasks under extreme conditions with limited consumables. To mitigate the risk of depleting critical consumable resources, we propose a wearable biomonitoring system that measures, analyzes, and predicts physiological signals, including heart rate, blood oxygen saturation, respiration rate, body temperature, step rate, and arm swing rate. Alongside barometric data, these metrics are collected by a suite of sensors integrated with a Nordic nRF52840 microcontroller that filters the information and transmits it via low-energy Bluetooth to a MongoDB database. Preliminary data is labeled and used to train a machine learning (ML) model. This model can then analyze new data, classify the activity, and make predictions about pure oxygen consumption. Finally, all data is sent to a desktop application that displays calculated biomarker data, predictions, and feedback. This system promotes autonomy by offering insights into physical strain and enabling data-driven decisions during EVA tasks.