Sunday, October 20, 2013

CubeSats to the Planets

I’ve seen the evolution of computers used as an analogy for the evolution of spacecraft.  Once, all computers were massive machines (called mainframes).  Then the evolution of electronics allowed the creation of what were known as mini-computers that expanded the number of organizations that could afford the computer.   Further evolution brought the personal computer to billions of people, and now the smart phone allows hundreds of millions of use to carry computers in our pockets.

Similarly, planetary spacecraft originally tended to be large missions (today called Flagships).  NASA’s New Frontiers and Discovery program enabled a series of lower cost missions that can be compared to mini-computers.  As I discussed in an August post, planetary SmallSats could become the equivalent of personal computers. 

Before the decade is out, I expect that we will have at least one planetary mission based on the equivalent of the smartphone, a CubeSat.


Comparison of masses and costs of missions with different classes of spacecraft.  Click on the image for a larger view.

CubeSats were invented a decade ago by researchers at the California Polytechnic State University (CalPoly) and Stanford University to create a standard for university-built spacecraft.  As the name suggests, the basic configuration is a cube.  At just 10 centimeters on a side (or 1unit (1U) in CubeSat parlance) and weighing approximately one kilogram, CubeSats depend on the electronics miniaturization that has enabled personal computers and smartphones.  If a research team needs a larger spacecraft, cubes can be stacked together to create larger volumes with standard configurations up to six units and even twelve units.  The small size of CubeSats allows them to be carried into orbit cheaply as secondary payloads on launches of larger Earth satellites.


Example of a 1U CubeSat.  Credit: Svobodat, Wikimedia Common

I wasn't able to find a definitive number for how many CubeSats have launched, but Wikipedia lists a number of satellites.  An industry has grown up that will supply standard parts that allow a CubeSat to be built as essentially a kit (see, for example, the CubeSat Kit website).

Given the success of Earth-orbiting CubeSats, it’s only natural that engineers and researchers want to use their design philosophy to explore beyond Earth.  There’s even a regular conference devoted to the topic.

I don’t doubt that CubeSats will study other worlds in the coming decade.  However, there are a number of challenges designers will have to overcome: Longer lifetimes (years instead of weeks or months), propulsion for trajectory changes and potentially entering orbit, power (especially further from the sun), communications from vastly further distances than Earth orbit, higher radiation outside Earth’s protective magnetosphere, and instruments to make meaningful measurements.  The technology to enable serious planetary CubeSat spacecraft is under development but not yet here.  .  (This presentation gives a good overview of technology in development while this presentation shows the vision for one approach to interplanetary CubeSats.)

The first CubeSat mission planned to leave Earth orbit is a NASA-funded project led by the Jet Propulsion Laboratory with several university partners.  Called INSPIRE (for Interplanetary NanoSpacecraft Pathfinder In a Relevant Environment), the twin spacecraft will each be 3.8 kg and 3U in size.  The mission goals are modest: travel 150,000,000 km from Earth (in an orbit near the Earth’s around the sun), enhance CubeSat design to successfully operate outside Earth orbit, and conduct scientific measurements with a magnetometer and simple camera, and communicate with the Deep Space network.  The spacecraft should be ready launch in 2014.


Inspire CubeSat design.  Acronyms: ACS - attitude control system; C&DH - command and data handling; EPS - electrical power system; UHF - ultra-high frequency.  Credit: JPL.


Each INSPIRE spacecraft will generate just 20 watts of power using body mounted solar arrays.  A downside to this approach is that at least half the spacecraft’s body is in shade at any given time, limiting power.  Several proposed designs would have more substantial arrays with pointing mechanisms that allow the arrays to be kept pointing at the sun.  One design would also use the solar panel as a communications reflector to boost data rates, helping to address the problem of deep space communication.  However, even with larger solar panels, power levels will still be modest.  One of the more capable proposed designs, by Tethers Unlimited, would generate 80 W in Earth orbit.  At Mars, the power levels would be ~40 W to power the spacecraft systems, instruments, and deep-space communication.  That would be a tight energy budget.

Perhaps the greatest creativity in the interplanetary CubeSat community is going into propulsion.  Even if a CubeSat simply flies by its target world, it would still need to perform several trajectory corrections along the way.  Entering orbit around another world or rendezvousing with an asteroid would require a more substantial propulsion system that would need to fit within a liter or two of space.  A challenge for engineers is that their fuel must be utterly safe to launch as an auxiliary payload with a primary spacecraft.  Carrying traditional rocket fuels that could become explosives wouldn’t be allowed.

At least two teams are proposing to carry water aloft and once in flight covert the water to hydrogen and oxygen, which are excellent traditional rocket fuels.  Other teams are developing ion engines that convert substances such as xenon into high speed plasma jets (the same basic technology on a much, much smaller scale as the Dawn spacecraft is using to reach the asteroids Vesta and Ceres).  A JPL team has proposed a 6×6 meter solar sail that would use the pressure of sunlight for propulsion.  Several teams are developing electrospray micropropulsion systems in which a liquid is drawn up tiny capillaries and electrically charged and expelled as high energy ions.  A version under development at JPL would simply melt tiny blocks of the element indium to provide the liquid, creating a system that is extremely simple and robust.


An example of one set of technologies and design for an interplanetary CubeSat.  Click on the image for a larger version.  Credit: JPL

Once an interplanetary CubeSat reaches its target, what measurements could it make?  Some instruments, such as magnetometers, already are small and can easily fit within a CubeSat spacecraft.  Other instruments, such as cameras can be made quite small.  The wide-angle JunoCam camera aboard the Juno spacecraft is 1.66 kg (not including the two kilograms of radiation shield for the Jovian environment).  (These were the best figures for these masses I could find, but may reflect early estimates.)

One proposed instrument would be a miniaturized version of an imaging spectrometer under development by JPL and Louisiana State University.  The design would both take pictures of the target world(s) and the spectral data collected would provide information on the surface composition.  While larger versions of the instrument exist, development is needed to reduce its size to fit within a CubeSat.

Where might independent interplanetary CubeSats go?  Given their limitation on fuel, power, and communications, I suspect most independent missions are likely to stay relatively close to Earth.  While some teams will undoubtedly do a dedicated mission to orbit Mars or Venus (to show it can be done if nothing else), many much more capable spacecraft have and will continue to visit those worlds.  However, CubeSats could be used to scout large numbers of near Earth asteroids both for science and to prospect for resources.  Tethers Unlimited has proposed a 6U, 9 kilogram CubeSat called HAMMERsat that could flyby and image 15 near Earth objects in a 2.5 year mission.  The company estimates that once the first HAMMERsat is designed and proven, subsequent copies could be flown and operated for approximately $3+M.    (The first spacecraft, however, could cost several times that figure I suspect.)

I think another likely destination will be our own moon.  While many larger spacecraft have and will visit this word, too, the moon is close by and well suited to CubeSat missions.  Just as many university teams have flown Earth-orbiting CubeSats, I suspect a number will also target our moon.


Microcosm's Hummingbird interplanetary SmallSat concept that would incorporate elements of CubeSat design in a larger (but by traditional planetary spacecraft standards, still quite small) spacecraft.  Credit: Microcosm. 

Another approach for independent missions would be to forego strict adherence to CubeSat form factors.  In this approach, CubeSat components would be used and augmented with larger, more capable components as appropriate.  Microcosm has taken this approach with their Hummingbird design.  It would carry a capable telescope that by itself would weigh over 3 kg (almost the weight of each INSPIRE CubeSat).  The attached instrument(s) could be up to 30 kg.  The spacecraft would carry sufficient fuel to enable complex missions.  (See my previous post for JPL and NASA concepts that also would scale up CubeSat technology to SmallSats.)

I believe, however, that the greatest potential for CubeSats may be as auxiliary spacecraft carried to their targets by larger, more traditional spacecraft.  These CubeSats can play any of several roles.  They could carry out measurements in locations where the mother craft can’t be.  Or they could work in coordination with the mother craft and/or possibly other CubeSats to make measurements in several or many locations at once.  And finally, they could be sent on “suicide” missions to locations too dangerous to send the mother craft.

As an example of the first possibility, consider a flyby mission to Uranus.  Each of Uranus’ five major moons appear to have diverse geologic histories and two are candidates for hosting their own internal oceans (see Oberon and  Titania) and Ariel has features like Saturn’s Dione suggesting internal processes that modified the surface.  As a mother craft transits the Uranus system, it could flyby only one moon closely enough to make detailed imaging, composition, magnetometer, and gravity measurements. To enable measurements in many places during the flyby, the mother craft could drop off multiple CubeSats that would perform their own close flybys of a moon and then relay their results back to the mother craft.


Uranus's moon Titania as seen in low resolution by the Voyager 2 spacecraft in 1986.  Credit: NASA

Researchers have proposed a number of possibilities for missions that would use Cubesats to make dispersed, coordinated measurements.  One proposal would have a mother craft and several Cubesats detect the radio waves characteristic of lightning at Venus and triangulate the location of the bolts.  If the lightning were located preferentially over volcanoes, that would be powerful evidence of active volcanism on Venus.  Another proposal, Atromos, would modify the CubeSat form factor to enable low cost weather stations on Mars to study processes such as dust storms from multiple locations simultaneously.  Another concept for Mars would use multiple CubeSats to form an orbital network for meteorological observation.  (This presentation briefly describes this concept and provides good of background on the challenges and possible solutions for planetary CubeSats.)  Multiple CubeSats could also deployed during a flyby of a moon, asteroid, or comet to, for example, image the world from multiple vantage points and enable high resolution gravity measurements via spacecraft-to-spacecraft radio tracking.

Many possibilities exist for CubeSats sent to locations too dangerous for a multi-hundred million dollar spacecraft.  One possibility would be for a Europa flyby mother craft to deploy CubeSats to crash onto the surface of Europa.  As the CubeSats make their plunge, they would gather high resolution images of potential future landing sites or areas of scientific interest that they would relay back to the mother craft.  Another possibility would be to have a CubeSat skim above Uranus on a flyby of that world for high precision gravity measurements, a trajectory that carries considerable risk because of that world’s rings.

While I believe that planetary CubeSats will play an important role in future exploration, it’s important to also recognize their limitations.  There are some missions where many spacecraft making simple measurements would enable studies a single spacecraft couldn’t do (see above).  With scientific payloads of a kilogram or two, however, the measurements CubeSats could make will be constrained.  Sending one or a hundred CubeSats to Europa, for example, would not equal the measurements that would be made by the 100+ kg of highly capable, synergistic instruments proposed for the Europa Clipper mission. 

Also, although CubeSats are cheap compared to traditional planetary spacecraft, their costs will still be meaningful.  If a CubeSat is intended as an integral part of a mission’s science goals, then it must have the same reliability as the science instruments on the mother craft.  Carrying a capable CubeSat plus its deployment device might add 15+ kg of mass to a mission (and multiply that figure for each additional CubeSat carried).  The CubeSat design, hardware, and testing would add several million dollars in cost.  The mother craft may have to be upgraded with a communications system and software to relay commands and data to and from the CubeSat at a cost of additional millions of dollars.  The entire system of CubeSat(s) and mother craft would have to be extensively tested.  In the end, deciding whether to carry a CubeSat as part of the core mission could come down to a tradeoff: carry an extra spectrometer or radar or another instrument on the main spacecraft or carry a CubeSat or two?


One possible role of planetary CubeSats, though, especially excites me.  In Earth orbit, CubeSats have opened space flight to university students around the world.  Perhaps future planetary missions could reserve the mass for one or two student-led CubeSats that would operate independently of the main spacecraft, limiting the cost impacts to the primary mission.  Imagine how the opportunity to build their own spacecraft to make measurements at Mars or a comet or wherever in the solar system could energize the next generation of planetary scientists and engineers.

9 comments:

  1. Monte Davis used to chastise commercial space advocates. The "space cadets" would point at the spectacular improvements in computers once they became commodities on the free market. They hoped for the same in space flight. Davis would point out humans and their life support aren't amenable to miniaturization. Nor, due to Tsiolkovsky's rocket equation, are the propellant, tanks, and rocket engines it takes to deliver humans to space.

    I would reply that Moore's law has everything to do with commercial space. That it would make more able, less massive and more compact prospecting probes as well as robots to close the business case.

    My predictions are ready being partially realized in the form of cube sats.

    Without Moore's Law, Planetary Resources wouldn't have a prayer of delivering a fleet of able Arkyd prospector probes to orbit.

    Telerobots and robots able enough to do lunar or asteroid mining are still a ways off but I believe they're coming.

    And if robots can mine and manufacture propellant, water and air, cover habs with radiation shielding and build other infrastructure, it would become much easier to send humans to the moon, asteroids and other destinations.

    I believe Moore's Law has everything to do with the possibility of exploiting space resources, Davis' complaints notwithstanding.

    ReplyDelete
  2. Yes!! Mr. Hollister David is right that without Moore's Law, Planetary Resources wouldn't have a prayer of delivering a fleet of able Arkyd prospector probes to orbit.

    Regards,
    Mangosteen Juice

    ReplyDelete
  3. In general I'm a strong proponent of NASA and progressive non-American space agency missions of all types, that require cutting edge LIDAR sensors, and cutting edges RADAR sensors of the Bands ranging from X to UWB (I think). I'm also for the development of automatic target recognition capabilities. Right now these use limited guestimates. In the future certain still-frame movies; the activities of people or robots will signal WMD activities.
    I believe visual mapping of the living areas and potential robotic environments, of all of Earth, should be accomplished in near real-time, to find who is R+D-ing WMDs. This should narrow down the search space and with wise (ATR) selection, should not represent too much of a privacy threat.
    Consumer electronics and some slowing down of R+D are the main threats if the administration is wisely educated/picked.
    I'm spamming your post here because CubeSats appear to be an ideal filter for making such a sensor network cost-effective.
    I suspect the (AI supercomputer) infrared sensors might be easily shielded from in the future.

    ReplyDelete
  4. ...to elaborate, the Control Theory wiki suggests activities related to R+D can be analyzed if they are pattern recognized from a still frames or video database. Then other sensors or police could investigate further. UV or stronger EM media sensors will cause cancers or burns, and infrared seems easy to shields; not bodies but computation.
    But both humans and robots will require motion to build WMDs.
    So I prefer space missions that use LIDAR and visible band radar frequencies, in the hopes these sensors made cheaper will be all that is needed to prevent harder to find robotic R+D from being developed. I'm annoyed NASA was shut down and a friend is mad FDA human trials were stopped.

    ReplyDelete
  5. Van,

    The cubesats do not need to travel with the mothership. This way only the communication hw needs to be modified.
    Take, for example, JUICE, arriving around Jupiter in the '30s. It's unlikely that ESA would let cubesats compromise or disturb the primary mission. However, when the primary misison is over and JUICE is around Ganymede, it could have spare time to work as a relay for cubesats arriving from earth.
    These could be easily launched from earth in small groups. The problem is more slowing them down when arriving at Jupiter.
    For that I remember reading about some very efficient and very slow transfer orbits between planets that allow to move through the solar system with minimal energy if you have time. Each cubesat would still need some propulsion system similar to the ones proposed.
    A very ow orbit around Europa, mapping the surface at high resolution would be a very good mission.

    ReplyDelete
  6. And, by the way, does Juno needs to be destroyed ?
    If it could be parked on a safe enough orbit, it could work as a cubesat relay.

    ReplyDelete
  7. Enzo -

    Juno does not have any communications relay capability that I'm aware of. I also don't believe that any is planned for JUICE.

    The challenge for CubeSats at Europa is that the spacecraft are small and can carry little radiation shielding. Lifetimes probably are in hours.

    The concept under discussion would carry CubeSats inside "vaults" on a mother ship for deployment just a few hours before impact. I don't think a CubeSat could enter Jovian orbit and pump down to do multiple Europa flybys and survive the radiation.

    ReplyDelete
  8. Both Juno and JUICE must be able to receive data from earth (sw updates, commands, etc.). They are also capable to send back information to earth. I was thinking of a transmitter on the cubesat using the same frequencies as the one used by earth but much lower power (given the smaller distance). So both probes should be able to act as a relay by receiving the data from the cubesats and sending it back to earth. It could even be a laser on the cubesat and the telescope or star tracker used as a receiver. I believe this has been shown for very high bit rate for LADEE (622 Mbits/s):
    http://www.nasa.gov/content/goddard/historic-demonstration-proves-laser-communication-possible/

    Given that the receiving telescope on JUICE won't be as good as the one LADEE sent data to on earth, I expect lower bitrate. Still, even a fraction of 622 Mbit/s would be great.
    For radiation I really don't know, but when I played with Wolfram Alpha the radiation shielding suggested wasn't absurd. See previous post.

    ReplyDelete