This work presents a modular battery management system (BMS) for space applications. Based on commercial off-the-shelf (COTS) components, the system provides state of charge (SoC) data for battery power budgeting and lifetime management. While intelligent BMS are common on Earth, space applications are constrained by severe computational limits. Therefore, the novelty of this work lies in embedding a linear Kalman filter (KF) for SoC estimation and an SoC-driven cell balancing algorithm entirely on a standard space-grade 8-bit microcontroller. This reliable on-board solution prevents faults and reduces the reliance on high-bandwidth Earth telemetry. The KF-based SoC estimation and balancing approach is validated in an electrical test campaign. The battery system is charged and discharged in more than 100 cycles representative of a space mission profile to evaluate the software. During the measurements, the SoC is determined by the algorithm implemented on the microcontroller and compared to a reference. A root-mean-square error (RMSE) of less than 1 % is obtained for the on-board SoC estimation across long-term tests with varying depths of discharge (DOD). Furthermore, the method proves its drift stability and correction capabilities in a long-term test with faulty sensor data. Balancing is achieved during all test cycles.
Eilenberger et al. (Mon,) studied this question.