Micro-Space >> Ultralight Manned Spaceflight

[rpspeck]

  SBIR PROPOSALS

Micro-Space recently submitted the following 10 NASA SBIR proposals.  None are “naked concepts”, but all build on substantial Micro-Space development work, often using technology embedded in our production products.     

A  =  Non Radiated Field Data Links for a MicroSat Launch Vehicle  (S4.01-9927)
 
B  =  Line Cutter Couplings for NanoSat Cluster Launchers  (S4.01-9182)

C  =  Automatic Solar and Celestial Navigation on the Moon and Mars  (O4.03-9181)

D  =  Near Vacuum Gas Ionizer to aid in  Lunar Dust Control  (X5.02-9180)

E  =   Quadrature Interferometric Video Holography  (A4.01-9179)

F  =   Mars Entry Long Range Atmospheric Temperature Profiling  (S5.01-9178)

G  =  RF Link Interferometry to guide PAV Orbital Rendezvous  (S5.04-9177)

H  =  Electromagnetic 6 DOF ADAC + Divert for Formations + Docking  (S5.01-9176)

J   =  Integrated Environmental CO2 + H2O Control with MEA Solution  (X2.02-9175)

K  =  All Temperature Paramagnetic Oxygen Sensor  (X2.01-9174)



SBIR contracts do not eliminate the need for outside funding, but compliment it.  Co-funding, from within NASA, the DOD, or outsiders,  is essential for any SBIR effort to avoid  a “dead end” and go on to produce useful products.  Earlier co-funding is necessary for SBIR contract “enhancement” or “Fast Track” status.  Since all such arrangements take time to develop and negotiate, the best efforts search for such support long before the SBIR funded work ends – optimally, before it even begins.  This announcement is an invitation for discussions of such cooperative arrangements.

Micro-Space will address these proposals in more detail in this forum.  The proposals fall roughly in to Five Pairs: 

A+B focus on enabling technologies for small launch vehicles.  They will complete the integration of our “Propulsion Modules” into First and Second Stage systems during enhanced Phase II work. 

G+H focus on Rendezvous and Docking technologies, particularly for small PAV (Planetary Ascent Vehicles), like the planned Micro-Space  “Lunar Sample Return” system, or NASA “Mars Sample Return”. Enhanced Phase II efforts would see both techniques demonstrated with orbital “CubeSats”. 

C+D cover Lunar Surface technologies, some of which may be essential for even GLXP Rovers.

J+K specifically reflect our long duration, human Life Support work.

E+F focus on optical “gaging” technologies.  E (Video Holography) is a laboratory technique , very useful in wind tunnels, while  F  would analyze the Mars atmosphere before Aerocapture or Entry. 

As usual, for Micro-Space, Phase II efforts would produce working prototypes of all these systems. 


Neither the  likelihood of funding for some of these proposals, nor the positive independent NASA response to discussions of these topics eliminates the need for additional investment or funding, although it does ratchet down the dollar levels, and leverage the results.   

[rpspeck]

MARS MISSION Hardware Progress

The Micro-Space “Centrifugal Evaporator”, for Zero G, Water Recycling has now completed 3 years and one week of continuous operation , at 14 times its required speed, in accelerated life tests.  Tomorrow, the total will be 1,111 days of run time, well beyond the total for a “Slow” (Minimum Energy) Mars mission.  As noted earlier, this compact unit is full size to process an adult human's liquid metabolic waste, saving  more than one kilogram a day of mission supplies.  This test will continue. 

Food required runs about ½ kg per day (with some high energy foods), and Oxygen at 640 grams a day without recycling.  Efficient CO2 capture will allow partial recapture of the Oxygen, with no net O2 input then required.  (Human metabolism produces H2O as a byproduct, which can be electrolyzed (in another of our prototypes) to yield Oxygen.  This more than makes up for normal reprocessing losses.)

Bulk water, like wash water, can be recycled by standard “Reverse Osmosis”– producing concentrated contaminant solutions for evaporation.

Our test of a “Transfer Membrane”, for CO2 capture into recycled MEA solutions, has now completed 18 months of run time, and is thus equally promising.  This is also a core components of a Zero G “Life Support” system. 

Both systems are sufficiently compact and low mass that spare systems and components are easily accommodated. These encouraging results support our prediction that “Low Cost”  Human Mars Missions (ca $100 Million budget) will need only 500 kg of consumables per traveler.

[rpspeck]

Good News for Space Research

NSF (the (US) National Science Foundation) has recently entered “Space Research”, initially funding high quality “CubeSat” efforts.  With its long history of supporting truly innovative research work by small teams, NSF support  is a very encouraging development. 

Less positive for American technology, but good for Space, India has announced at least 6 more orbital  launches for CubeSats and MicroSats up to 100 kg. Several of these flights are for “Sun Synchronous” orbits,  which are very desirable for Earth Linked satellites: those involving Earth Sensing or communications. The Universities of  India, so successful with turning out programmers to capture a good share of the software development market, will have somewhat of an inside track, working with the nearby launch center on the South India coast.

No comparable launch opportunities have been announced by American or European Launch providers.

As I have mentioned earlier, these launches are capable of starting a “Google Lunar X PRIZE” competitor on its way.  One hundred kg in LEO can translate to 10 kg of landed mass on the Moon, and this is sufficient for the Google Prize.  Propulsion system hazards will always be an issue, but these systems are not out of the question for the India launches.

CubeSats have been specifically identified for  full operational demonstrations of two of the technologies in the recent Micro-Space NASA SBIR proposals.  There is no question that with the modest funding outlined in these proposals, innovative Rendezvous and Docking technologies will be successfully demonstrated.

[rpspeck]

Navigation on the Moon

As promised earlier, I will discuss the Micro-Space SBIR (Small Business Innovative Research) proposals in this forum. I will discuss our C  Proposal, “ Automatic Solar and Celestial Navigation on the Moon and Mars  (O4.03-9181)” today.  This has direct application to Lunar Rovers – particularly those rolling or hopping considerable distances – and even more application to human explorers.  Only the simplest component is likely to be needed for GLXP (Google Lunar Rover) systems with their 500 meter travel, but that system is still very desirable. 

It is easy to take for granted both the incredible performance of GPS and our classic road signs, which combine with good maps to make navigation fairly easy.  Traveling where signs don't exist has always been “interesting”, without today's GPS data, and neither exists on the Moon or Mars.  In theory, a GPS type cluster of satellites could be  maintained around either body to provide GPS type data, but I am always talking about the first pioneers, not much later travelers, and these satellite clusters and their support  systems are very expensive to install and operate. 

Near term visitors to the Moon (or Mars) will find their navigational aids set back not just a century (with no GPS or road signs), but several millennia, since a “Magnetic Compass” is also useless: neither body, or any inner planet of the Solar system other than Earth, has a usable magnetic field . Fortunately, if the sun is “up”, it is usually visible on both bodies, so that “Celestial Navigation” has at least one good optical reference.  Gyros, and particularly tiny, low cost MEMS units, have significant drift and can not provide a good directional reference for more than a few minutes. Even “Deductive Reckoning” needs a good  directional reference to convert estimated motions into position changes. 

Unless “Back Tracking” is totally reliable, a good directional reference must be maintained.  Apollo operations on the Moon left highly visible tracks, but over rocky surfaces on Mars, tracks  have not always been so visible.  And, in a few hours, crisscrossing tracks can become dangerously confusing!

It would not be a good idea to operate even a short range, GLXP rover without a sustained directional reference so that planned travel can be closely approximated, and a clear line of sight to the lander need not be maintained. Small hills and dips, as well as radio “Surface Reflections” can severely degrade  operations which depend on an uninterrupted link even over fairly smooth terrain, let alone the unexplored rough Lunar terrain which will be of most interest to follow on customers.

A simple Sun Sensor, with a few sensing elements and analog signal processing, can be sufficient for better than one degree absolute direction, provided that the travel history and clock time are known and the Sun Sensor can be kept level.  A tilt sensor, optimally augmented by tiny MEMS gyros, allows a “Strap Down” system to work even better than one with moving gimbals and can work while the rover travels.

Six well matched and calibrated elements, for example, with good linearity, temperature compensation and low noise can cover the required 360 degree field of view with a fraction of a degree resolution.  (This ignores the awkward  “solar zenith” zone, with its +127 C lunar surface temperature at latitudes where it can occur.)  Alternatively, 1000 low resolution “binary” light sensors can provide comparable resolution.  The optical sensors must be able to monitor considerably more than the 2 Pi steradian  upper hemisphere to accommodate low sun angles viewed by a tilted optical sensor.

The apparent sun direction is shifted by the tangent of the sun altitude angle times the roll angle perpendicular to the sun direction, for small offsets, with no first order effect from pitch errors along that axis.  (The tangent factor of course “blows up” near zenith.)  Both active and passive “levelers” can be used to remove these effects, but moving mechanisms are to be avoided whenever possible on a lunar craft and in any case have undesirable responses to motion transients.  A “Strap Down” system, with no moving parts, is preferred and the computer data processing required is not a great burden. 

Adding a  tilt sensor (2 axis accelerometer), and tiny MEMS gyros, will produce both excellent short term directional information (from the gyro), and excellent long term direction – using averaged data from the optical sensors.  These sensors mass less than a gram and eliminate all moving parts.  Wheel motion data provides a linear velocity for the rover and is used for “Deductive Reckoning”. (Wheel slippage creates a navigational problem.) Acceleration offsets of the measured “plumb line” will be computed from this wheel motion data.  Offsets due to the “centrifugal pseudo force”  are computed as the cross product of the linear velocity and the vehicles angular rotation (radians per second) as it turns.   

     

Determining the absolute position  on the Moon (or Mars), to eliminate the errors that build up in “Ded Reckoning”, is more complicated, as is navigation at night or on the Lunar “Far Side”.  Providing these with the same, trouble free optical package as the “Direction Reference”, taps more of the special expertise of Micro-Space, is the focus on our SBIR Proposal, and will be discussed in more detail another day.           

         

 
NOTE:  The SBIR program grants are officially, and very commonly in practice, “Dead Ended”.  Although there is often an extension from Phase I “Feasibility Study” to Phase II,  “Operational  Breadboards”,  the so called Phase III continuation – including production, for interested customers - must be funded through other private or agency channels.  So called “SBIR Mills” specialize in completing the funded research , filing the required reports, and moving on to other topics with little expectation of a follow on.  This is not the history of Micro-Space and its affiliates, which have delivered large quantities of “High Tech” products with little outside funding for research.     

Developing NASA or Commercial customers for our proposed technologies will take some time, thus it is very important to start this process early, through a variety of marketing efforts: these discussions are one components of that effort.  We welcome discussions with customers who could benefit from these technologies, and in every case could begin delivery of “Beta Test” hardware before the end of the SBIR projects.   

[rpspeck]

1/2 Gram Video Camera

We are working at the moment to integrate a ½ Gram (!) Video Camera (with near Broadcast image quality) into our 25 gram “Space Probe”.  We have an adequate, similar mass video transmitter.  This probe was originally planned as an “FAA Exempt”, “Weather Probe”, capable of being guided into a selected portion of a tornado or other violent storm system.   With no customers for this experimental system, we are reconfiguring it as a “PicoSat” or Pico Spacecraft.  It will probably be the first of our systems to approach the Moon, as less than a kilogram in LEO would allow acceleration of this guided probe to Lunar Orbit.

Video sensors massing no more than a gram are capable of delivering the GLXP required HDTV images.  The required optics will weigh much more than this, but possibly still only a few grams.  As mentioned earlier, the minimum mass necessary to WIN the Google Lunar X PRIZE has yet to be determined! 

The ½ gram camera is certainly able to serve as a sun and star tracker, with a small cluster covering the entire sky.  I will return to the navigation topic soon, but have been working to integrate images with these BLOG texts, since that makes them more interesting and understandable. 

It has been noted by some that existing “Star Trackers” could serve for Lunar navigation, and this is true.  However, the “Proven” units mass 2 to 5 kg, use 5 to 20 watts of power, and have significant stabilization and glare problems.  (Data from “SMAD”, Third Edition).  These are not things one wants to mount on top of a helmet!  More pointedly, these are expensive enough that they are not used without very good reasons, and have never been used on very small satellites (like CubeSats, whose total mass is less than the listed star sensors.) 

It is a good idea to rethink the systems we need to use, because the “Proven” units may be far from optimum and may even hide operational realities behind old habits.  I won't consider using the 8088 processor for my spacecraft (the engine of the 1981 IBM PC and the Space Shuttle's main computer).  I prefer the fractional gram, up to 50 MIPS, low power processors I am using.

[dunderwood]

However, the “Proven” units mass 2 to 5 kg, use 5 to 20 watts of power, and have significant stabilization and glare problems.  (Data from “SMAD”, Third Edition).

SMAD (at least my copy of the 3rd edition) has a copyright date of 1999.  That's quite some time.  I'm sure proven start trackers are much more efficient by now.

[tnphysics]

Nowdays it could probably be done with a single Pentium processor for computation.  The other components required are the already-needed camera and some motors.

[hop]

However, the “Proven” units mass 2 to 5 kg, use 5 to 20 watts of power, and have significant stabilization and glare problems.  (Data from “SMAD”, Third Edition).

SMAD (at least my copy of the 3rd edition) has a copyright date of 1999.  That's quite some time.  I'm sure proven start trackers are much more efficient by now.

Indeed http://www.aeroastro.com/components/star_tracker
300 grams, < 2 watts, off the shelf.

[Comga]

... However, the “Proven” units mass 2 to 5 kg, use 5 to 20 watts of power, and have significant stabilization and glare problems.  (Data from “SMAD”, Third Edition). ...

Current star trackers such as those presented in
http://www.dlr.de/iaa.symp/Portaldata/49/Resources/dokumente/archiv4/IAA-B4-0706P.pdf
are considerably smaller, 1kg and 8W for the flight proven system and 0.4kg and 2W for the advanced system.  Another company
http://www.spacedaily.com/reports/ISIS_To_Develop_Star_Tracker_For_Nanosatellites_999.html
has an SBIR grant to develop one half as large as that "advanced" system.

Not yet in the single gram regime though.

[hop]

Not yet in the single gram regime though.

Of course, the sensor is only a relatively small part of the whole tracker. Optics and housings account for a lot.

[Comga]

Those masses are totals, with electronics, sensors, cables, and optics, even 15g brackets to hold the heads.  IIRC, the heads for the the ASC are 60 grams apiece, optics and sensor. 

[hop]

Those masses are totals, with electronics, sensors, cables, and optics, even 15g brackets to hold the heads.  IIRC, the heads for the the ASC are 60 grams apiece, optics and sensor. 

Yep, understood. What I was trying to get at is that a rpspecks proposed star tracker based on his 0.5g camera could very well end up with a similar mass to the existing state of the art mini star trackers.

[Comga]

Good point.  None of the 1kg, 0.4kg, and what may be an 0.25 kg systems are dominated by their sensor heads. 

[rpspeck]

As noted by others in Forum responses to my  discussion of low mass “Star Trackers”,  progress has been made by established suppliers for these units.

But note that my ½ gram CAMERA unit INCLUDES its housing and optics.  Admittedly a good computer may also weigh ½ gram, although a minimal computer weighs less than 0.01 gram. Brackets, and similar parts tend to have a mass which is proportional to the mass of the assembly they hold. 

My point is that our assumptions – even mine – are quickly becoming our of date.  A new generation of experimental spacecraft and satellites are going to make “Yesterday's” standards look pathetic.  The one kilogram CubeSat is soon going to become the equivalent of an old “Low Cost, Small Satellite”  for developmental work with similar technical capability (at about 1/100 of the launch cost),  and the MicroSat is going to become a unit with mind boggling performance! 

I have been doing some serious modeling for active array antennas, using low cost commercial components, on such a MicroSatellite, and the focused RF, multi subscriber capability makes satellite “Narrowcasting” quite practical!

[Lampyridae]

Nowdays it could probably be done with a single Pentium processor for computation.  The other components required are the already-needed camera and some motors.

Standard electronics won't cut it out in space, especially for critical systems. Rad-hard electronics must be used (but I suspect you already know this).

I suppose for a cubesat within the Van Allen belts that'd be OK though.

[hop]

This masters thesis might be of interest http://dspace.mit.edu/handle/1721.1/36170

But note that my ½ gram CAMERA unit INCLUDES its housing and optics.

But is it remotely comparable to those of actual working star trackers ?

Because there are frequently cheaper ways of getting coarse attitude measurements (e.g. sun sensors, magnetometers, earth sensors...) most real star trackers are pretty precise. This requires quality optics and decent light gathering capability.

What's the maximum magnitude star your camera can detect, in a reasonable exposure* without being drowned out by noise** ?

* if your attitude is changing too fast relative to your exposure, you're star tracker will just see trails.

** cheap, miniature sensors tend to be noisy. Many consumer digital cameras only take acceptable pictures thanks to extreme noise reduction, and even so struggle to capture stars.

Will your housing keep the electronics and optics in an acceptable operating environment ?

How well do your sensor and processor tolerate radiation ? Some COTS stuff does OK, but miniaturization tends to have a negative impact here.

Do the optics correct for field curvature ?

You are once again implying that those who actually work in the field are ignorant or incompetent, while at the same time demonstrating beyond all doubt that you haven't even done cursory research. This is a pattern we've seen before, and it does you no favors.

[rpspeck]

You are once again implying that those who actually work in the field are ignorant or incompetent, while at the same time demonstrating beyond all doubt that you haven't even done cursory research. This is a pattern we've seen before, and it does you no favors.

My point ... as indicated above ... can be rephrased as “See the Handwriting on the Wall!”   

You assume that this short note covers all my experience with optical sensors and thus you can “puff” about “The Real Problems”.   Since I lead the development, production and calibration of 40 high resolution, optical HMD and HUD automatic test and calibration systems (delivered to DOD components) with ½ “second of arc” resolution, I do know something about “The Real Problems” and their solution!  I also have been involved with image sensors since the days of the “Iconoscope” (with its electron gun angled into the larger, “target chamber” from the front).  I am aware from experience - not just theory – that there is a steady evolution in these technologies.  The newest versions always bring new problems which demand new solutions – and possibly a focus on other applications!  The new applications made possible by such progress (even with initially lower performance) are awesome!

I did not suggest that this ½ gram unit would replace the best existing star trackers.  But I did suggest that it opens up applications never before possible for star trackers in space, for mass, power and cost reasons.

For many of these “Limiting Magnitude” is not a problem.  With 20 stars – and five planets – brighter than magnitude 1.31, a nice group is available at any time.   Noise is always a problem, particularly with the “Sub Pixel Processing” methods we use to obtain extreme resolution with significant field of view. Solutions exist – which I won't discuss in detail. 

Your comments and attitude match the 1979 comments of IBM “Pros” concerning “Toy Desktop Computers”. (Since called “Personal Computers”, PCs)   I don't think I will change YOUR attitude.

But for the students and young engineers reading this, think about the PC computer history, and its likely parallel in Spaceflight, before you plan your career!

[rpspeck]

Visible Lunar Lander Progress

I was able to attend the Northrup Grumman Lunar Lander Competition in Las Cruses, New Mexico Last week.  It was good to see both completion of one Round Trip (90 second legs) flight profile, and to see a second team in the air with a FAA licensed flight!  “TrueZer0” (TrueHero!) helped viewers understand that keeping a hovering rocket in flight is a lot harder than Armadillo makes it look!

Armadillo didn't exhaust their deep well of  “bad luck”, but did put two adequate 90 second flights together to win $350,000.  Two years of “Coming Close” has been frustrating for both Armadillo and viewers, but it also erased Armadillo's multi year head start, flying this type vehicle, and eliminates that as an excuse for not flying competitive vehicles.  One other team – TrueZer0 - finally did!

Micro-Space fully intends to fly in future events, and can't conceive of any GLXP effort without NASA sized (Billion Dollar) funding succeeding on the Moon without successful lander tests on the Earth. 

The 180 second hover duration is of course required for a Moon landing, but the 25 kilogram payload definitely is not needed to be a GLXP winner.  I envision Lunar Lander competitions – beyond the NASA funded prizes – growing smaller in mass as well as emphasizing storable fuels (not LOX).  A 100 pound system in LEO (a small fraction of a Falcon 1, or PSLV (India) launcher payload) could be orbited for $1 Million and land 10 pounds on the Moon.  Given demonstrations of a reliable landing system (in New Mexico), the 20:1 Prize Payoff would be an interesting investment even without sponsorship.  A steady ramp up of CubeSat successes is proving the feasibility of low mass space systems.

[A_M_Swallow]

The 180 second hover duration is of course required for a Moon landing, but the 25 kilogram payload definitely is not needed to be a GLXP winner.  I envision Lunar Lander competitions – beyond the NASA funded prizes – growing smaller in mass as well as emphasizing storable fuels (not LOX).  A 100 pound system in LEO (a small fraction of a Falcon 1, or PSLV (India) launcher payload) could be orbited for $1 Million and land 10 pounds on the Moon.  Given demonstrations of a reliable landing system (in New Mexico), the 20:1 Prize Payoff would be an interesting investment even without sponsorship.  A steady ramp up of CubeSat successes is proving the feasibility of low mass space systems.

Currently SpaceX does not advertise a release system for multiple CubeSats on the Falcon 1 so Micro-Space would have to design one and find 7 other CubeSat customers.

NASA made a 3 micro-satellite support structure for Pegasus as part of its ST-5 mission.
http://www.nasa.gov/mission_pages/st-5/main/index.html

[dmc6960]

I thought Malaysia designed one and is flying it on the next Falcon 1 launch.

Currently SpaceX does not advertise a release system for multiple CubeSats on the Falcon 1 so Micro-Space would have to design one and find 7 other CubeSat customers.