Ethiopian Review

THE trouble with space probes is that once they have been launched their mission cannot be changed. But a test satellite planned for 2012 could change that: its flight computer will contain electronic hardware that can be completely reconfigured in space, allowing it to switch from, say, an atmospheric pollution sensor to a near-Earth asteroid detector.

Dubbed the “flying laptop”, the spacecraft is the brainchild of Toshinori Kuwahara of the Institute of Space Systems at the University of Stuttgart, Germany, whose idea will appear in a forthcoming edition of the spaceflight journal Acta Astronautica (DOI: 10.1016/j.actaastro.2009.04.011). The craft will carry a host of instruments and sensors, such as cameras, multispectral imagers, thermal infrared imagers, star trackers, GPS receivers and sea-surface-height sensing radar.

Using such devices in different combinations requires different computing configurations, both for operating them and for processing the data they collect. For this reason, standard probes are confined to one or two tasks. To enable his craft to multitask, Kuwahara will eschew the standard microprocessors used to process data in today’s flight computers. Instead, the flying laptop will construct the best possible microprocessor for the task at hand, using microchips called field-programmable gate arrays.

FPGAs contain logic gates that can be connected and disconnected by programmable switches. All that’s needed to move from one task to another is to retrieve the relevant logic gate connection settings from the flying laptop’s memory- or beam them to the spacecraft.

Kuwahara has to find a way to protect the FPGA circuits from charged cosmic ray particles, which can interfere with digital data and cause programming errors. He plans to use multiple back-up FPGAs all doing the same job at the same time, along with a program to decide which ones are performing correctly.

Kuwahara senses a commercial opportunity: the spacecraft could be rented out to different groups of researchers during the same mission, he says, spreading the cost of space-based research. The teams would be able to reprogram the probe to do the task of their choice.

By Paul Marks | newscientist