Part of the sky-high cost of running missions in space is down to the bespoke computer hardware on satellites and spacecraft. Space-bound computers are usually orders of magnitude less powerful than consumer hardware, but many times more expensive.
Now UK researchers want to knock several zeros off the price of space-faring computers by replacing today’s custom-built devices with the likes of smartphones and video games controllers.
Researchers from the Surrey Space Centre (SSC), part of the University of Surrey, and spin-out company Surrey Satellite Technology Ltd (SSTL) are running two projects to study how consumer tech fares in space.
STRaND (Surrey Training, Research and Nanosatellite Demonstrator) 1 will study how a Google Nexus One smartphone operates in orbit and whether it can be used to control a CubeSat satellite, a nano-satellite which weighs as little as four kilograms. The second project STRaND 2 will attempt to use Kinect, Microsoft’s controller for its Xbox games console, to dock two CubeSat satellites.
Dr Chris Bridges of the SSC and one of the STRaND project managers, said: “It’s trying to think outside the box and teach people that you can still be innovative at low cost.
“Instead of having to pay a lot of money for a bespoke piece of hardware, we are asking if you can go out and buy a $300 phone and use that.”
Part of the reason that processors inside modern smartphones are many times faster than those used in space-bound computer systems, is researchers tend to favour older hardware that has worked on space missions in the past.
“Technology has moved on and the aerospace community is a good number of years behind what we currently have. These smartphone processors are much lower power, faster and designed to be much more capable,” said Bridges. For example, the low-cost GomSpace NanoMind computer system that will be used in the STRaND-1 CubeSat satellite, for example, relies on a 40 MHz ARM 7 processor, compared to the 1 GHz chip inside the Nexus One phone.
“We’re trying to leverage this state of the art technology and see whether we can spin that into applications in space.”
The Nexus One phone will interface with the satellite’s computer systems via an I2C and USB board and Java-based middleware, which translates data and commands.
One of the biggest challenges will be ensuring that the phone works in orbit, where temperatures fluctuate between -20C and 80C and the handset will be bombarded by high energy radiation that can cause software to crash or behave in unexpected ways. There is also the question of how well its processor will work in a vacuum, without air to convect away heat from its surface.
To prepare the phone for the journey, researchers are simulating the thermal, radiation, vacuum and vibration conditions in orbit and studying how the phone behaves. Once they identify problems, they customise the phone’s Android 2.3.4 OS and software in an attempt to get it to run in a way that lets the phone avoid or recover from those problems.
“Ruggedising the phone requires, seeing what parts are vulnerable and whether or not you can protect them,” said Bridges.
The university has been ruggedising consumer technology for more than 30 years, in 1981 it customised a Speak & Spell so that it was able to work on board satellites.
If the phone works in orbit then it may be assigned limited control of the satellite. The team are also keen to test whether the low-cost sensors built into the phone – the accelerometers and magnetometer that help track its movement and orientation, its camera and microphone – can match or surpass the performance of the expensive custom-built sensors found on-board commercial and government satellites.
To measure the performance of the phone’s sensors the phone will run several apps – such as iTesa that will record the Earth’s magnetic field, two apps that will work out the satellite’s position by taking shots of the earth using the phone’s camera and Scream in Space, which will use the phone’s microphone to test the famous tag-line from the film Alien ‘In space, no-one can hear you scream’.
The prevalence of apps that could be customised to carry out tasks during space missions and the plethora of developers making apps for Android and Kinect is another reason pushing these consumer technologies into final frontier, he added.
“Potentially these are the types of apps that could one day find their way onto mainstream satellite systems,” said Bridges.
STRaND 2, will extend the idea of using consumer tech in space by fitting satellites with technology from the Kinect, Microsoft’s controller for the Xbox games console that tracks body movements using an traditional RGB camera and infrared sensors, allowing it to map objects in 3D space.
The mission will investigate whether two CubeSat satellites can dock when using Kinect components to track the 3D space around them. If it works it will be the first demonstration of a low-cost space docking system. Such systems are usually reserved for big budget space missions to the International Space Station.
“From our preliminary tests, it looks like the Kinect is a viable option but we still need to do a lot of ground testing on it,” said Bridges.
The team will have to develop custom drivers and software to get the Kinect running with satellite sub-systems. They also won’t be able to use their usual GomSpace satellite computer system, as it lacks the muscle to run Kinect, and are instead considering using an Android phone to run it instead.
“We will try to simulate as much as possible on the ground and figure out how can we hook this up to flight computers and get them to run autonomously,” said Bridges.
The STRaND team sees the relatively low cost but high powered nano-satellites they will test in these missions as intelligent “space building blocks”, which could be stacked together and reconfigured to build larger modular spacecraft or as mission objectives change.
The STRaND 1 satellite is due to launch this year, and Bridges believes that once consumer tech is proven to be viable in space, its low-cost capabilities will be a powerful motivator for major space agencies and companies to adopt it.
“As soon as we can fly these new systems, and get people more serious about it, there’s no reason why it couldn’t happen.”