Machine learning for predicting thermal power consumption of the
Mars Express Spacecraft

The thermal subsystem of the Mars Express (MEX) spacecraft keeps the on-board equipment within its pre-defined operating temperatures range. To plan and optimize the scientific operations of MEX, its operators need to estimate in advance, as accurately as possible, the power consumption of the thermal subsystem. The remaining power can then be allocated for scientific purposes.

We present several benchmark data sets for performance testing of machine learning methods. The data sets relate to multi-target regression tasks with 33 target numerical features. While the data sets relate to the same tasks, they are constructed on different time-resolutions thus differ in the number of examples ranging from ~65K examples in the lowest resolution to ~4M examples in the data set with the highest resolution. Therefore, these scenarios render various benchmarks for performance analysis.

The task at hand refers to predicting thermal power consumption of the Mars Express Spacecraft i.e. The Mars Express Challenge. Mars Express Spacecraft (MEX) has been orbiting Mars or more than 15 years, providing evidence of the presence of water above and below the surface of the planet, an ample amount of three-dimensional renders of the surface as well as the most complete map of the chemical composition of Mars’s atmosphere. The Mars Express Orbiter is operated by the European Space Agency from its operations centre (Darmstadt, Germany) where all the telemetry data is analyzed by the Advanced Concepts Team (ESA/ESTEC) in collaboration with the Data Analytics Team of the Advanced Mission Concepts Section (ESA/ESOC). The health status of the spacecraft is carefully monitored to plan future science observations and to avoid power shortages.

MEX is powered by electricity generated by its solar arrays and stored in batteries to be used during the eclipse periods. The scientific payload of the MEX consists of seven instruments that provide global coverage of the planet’s surface, subsurface and atmosphere. The instruments and on-board equipment have to be kept within their operating temperature ranges, spanning from room temperature for some instruments, to temperatures as low as –180°C for others. In order to maintain these predefined operating temperatures, the spacecraft is equipped with an autonomous thermal system composed of 33 heater lines as well as coolers. The thermal system, together with the platform units, consumes a significant amount of the total generated electric power, leaving a fraction to be used for science operations.

In sum, the main task is: Given three Martian years of telemetry data (August 22, 2008 to April 14, 2014) efficiently construct predictive model that accurately predicts the values of the electric current through the 33 thermal power consumers for the subsequent Martian year (April 14, 2014 to March 1, 2016).

Cite as:
 @ARTICLE{Petkovic2019,
  author={Matej Petkovi\'{c} and Redouane Boumghar 
  and Martin Breskvar and Sa\v{s}o D\v{z}eroski 
  and Dragi Kocev and Jurica Levati\'{c} 
  and Luke Lucas and Alja\v{z} Osojnik
  and Bernard \v{Z}enko and Nikola Simidjievski},
  journal={IEEE Aerospace and Electronic Systems Magazine},
  title={Machine learning for predicting thermal power consumption 
  of the Mars Express Spacecraft},
  year={2019},
  DOI={10.1109/MAES.2019.2915456},
  volume={34},
  number={7}}

Publications

  • Petkovic, M et al. (2019) Machine learning forpredicting thermal power consumption of the mars express spacecraft. IEEE Aerospace and Electronic Systems Magazine, 34(6)
  • Breskvar, M. et al. (2017) Predicting thermal power-consumption of the mars express satellite with machine learning. In Proceedings of SMC-IT 2017, pages 88–93.