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 a machine learning pipeline for efficiently constructing accurate predictive models for predicting the power of the thermal subsystem on board MEX. In particular, we employ state-of-the-art feature engineering approaches for transforming raw telemetry data, in turn used for constructing accurate models with different state-of-the-art machine learning methods. We Accepted at IEEE Aerospace and Electronic Systems Magazine 2 show that the proposed pipeline considerably improve our previous (competition-winning) work in terms of time efficiency and predictive performance. Moreover, while achieving superior predictive performance, the constructed models also provide important insight into the spacecraft's behavior, allowing for further analyses and optimal planning of MEX's operation. Index Terms machine learning, Mars Express spacecraft, ensemble learning, predictive modeling, random forest, gradient boosting, feature engineering I. INTRODUCTION M ARS EXPRESS (MEX), a spacecraft operated by the European Space Agency (ESA), is Europe's first spacecraft that orbits Mars. During its science operations, since the beginning of 2004, it has provided evidence of the presence of water above and below the surface of the planet [1], 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 [2].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.Predicting the power consumption of the thermal system is a non-trivial task. However, due to the aging of the spacecraft and the decaying capacity of its batteries, it is a very crucial one for optimal planning and execution of science operations on MEX. The power consumption is a dynamic process that changes through time, depending on various external and internal factors, such as long-term exposure of the spacecraft to the Sun or heat generated by the onboard instruments. F...
This paper presents the results of the potential application of machine learning techniques, specifically the Random Forest method, to spacecraft operations optimization. The test subject is ESAs INTEGRAL gamma ray observatory with the goal of demonstrating that AI techniques can reliably model the radiation environment of the satellite as it orbits the Earth and passes through the Earths trapped radiation zones in the Van Allen belts. The results clearly demonstrate that machine learning can approximate predictions of complex and dynamic radiation environment within +/-10% provided that an extensive data set is available and is adequately engineered. The consequences of such accurate data-driven predictions are that comprehensive physical models may be, under certain circumstances, an unnecessarily complicated solution to the optimization of scientific operations of Earth orbiting satellites.
This paper shows how the recent development and integration of a machine learning predictive model provided MEX with a more sensitive and accurate model of the thermal power consumption with points every hour. Machine learning (ML) techniques are highly applicable to multivariate problems with complex relations and constraints, such as spacecraft operations, particularly a context as complex as interplanetary operations. The thermal power consumption is the result of complex and dynamic system. A comprehensive understanding is difficult to achieve, due to these complexities. The approach is thus to understand the link between flight scenarios and the thermal power consumption on the 33 thermal lines identified as main lines of the thermal power subsystem. The insight given by this understanding helps building better models and finding the right points of improvement to make them ready for operations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.