Intelligent control system for life support and energy management in lunar habitats using TinyML and dual microcontrollers

Rojas J.C.T.; Perez L.Z.M.; Camacho M.R.; Cevallos O.O.; Huaman J.H.R. · Proceedings of the International Astronautical Congress, IAC (2025) · DOI: 10.52202/083078-0036

Resumen

The development of lunar habitats requires autonomous and low-consumption systems that maintain internal environmental stability with minimal human intervention. The paper, conceived at the National University of San Marcos in Peru, presents the design of a prototype intelligent system for energy control and life support in lunar habitats, based on a dual microcontroller architecture. One microcontroller is responsible for acquiring the internal environmental parameters of the lunar habitat. such as air quality, humidity, and temperature, using COTS sensors, as well as controlling actuators. In parallel, the STM32 performs autonomous inference through TinyML to recognize environmental deterioration patterns and activate adaptive measures without an Internet connection. The results show adequate energy performance and enhanced fault resilience, validated through physical simulations and software. This modular, economical, and intelligent solution can reinforce the robustness of a lunar habitat in technical compartments.

Instituciones: Universidad Nacional Mayor de San Marcos