History > 2006 > USA > Space (I)

No, it's not an alien.

Engineers are designing spacecraft that can think for themselves,

then execute their own commands,

without waiting for cues from ground controllers.

Feric

 Intelligent Beings in Space!        NYT        30.5.2006

http://www.nytimes.com/2006/05/30/science/space/30rock.html

Intelligent Beings in Space!

May 30, 2006
The New York Times
By KENNETH CHANG

A future space mission to Titan, Saturn's intriguing moon enveloped in clouds, might deploy a blimp to float around the thick atmosphere and survey the sand dunes and carved valleys below.

But the blimp's ability to communicate would be limited. A message would take about an hour and a half to travel more than 800 million miles to Earth, and any response would take another hour and a half to get to Titan.

Three hours would be a long time to wait if the message were: "Help! I'm caught in a downdraft. What do I do?" Or if the blimp were to spot something unusual — an eruption of an ice volcano — it might have drifted away before it received the command to take a closer look. The eruption may also have ended by then.

Until recently, interplanetary robotic explorers have largely been marionettes of mission controllers back on Earth. The controllers sent instructions, and the spacecraft diligently executed them.

But as missions go farther and become more ambitious, long-distance puppetry becomes less and less practical. If dumb spacecraft will not work, the answer is to make them smarter. Artificial intelligence will increasingly give spacecraft the ability to think for themselves.

"These technologies are already in operation on specific missions," said Steve Chien, a computer scientist who heads the artificial intelligence group at NASA's Jet Propulsion Laboratory in Pasadena, Calif. Scientists discussed some of the recent progress last week at a meeting of the American Geophysical Union in Baltimore.

HAL, the soulful conversationalist at the helm of the spaceship in "2001: A Space Odyssey," is not on the drawing board. The work so far has been more along the lines of Roomba, the robotic vacuum cleaner, with autonomy to perform certain specific tasks.

Dr. Chien's group wrote the software that manages the schedule of Earth Observing-1, a satellite that looks for natural disasters like volcanic eruptions, wildfires and floods.

The satellite, known as E.O.-1 for short, takes repeated pictures of the areas it watches, looking for changes that would indicate an eruption or other event. Other satellites or sensors on the ground can also dispatch an alert to E.O.-1, telling it of something it should look at.

"Almost immediately, within a matter of hours, the spacecraft is reprogramming itself to image these targets," Dr. Chien said, "and we can get rapid response imagery of breaking science events."

The spacecraft adds the new observation to its schedule and starts rearranging its other tasks. An observation that had been planned may be canceled or moved. All this occurs without a human in the loop. E.O.-1 just drops a note to the operators about what it has done.

When E.O.-1 was launched in 2000, people on the ground did the satellite's planning. The planning software was first tried in 2003, and the satellite now uses it full time. That not only sped up its reaction time, but it also cut its operating cost of $3.6 million a year by more than one-quarter.

Similar programming can be used for future planetary missions, perhaps for the next visit to Jupiter and its moons, to detect volcanic eruptions on Io or a shift in the fractures on the frozen surface of Europa.

NASA's two rovers now on Mars — the Spirit and the Opportunity — also possess a measure of thinking ability. As they drive, the rovers use stereo cameras to judge the distance and size of rocks in their paths in order to figure out how to maneuver around obstacles.

"At every step, it looks at dozens of potential choices," said Mark W. Maimone, a member of the team working on the rovers' software. "It picks the safest path that gets it closer to its goal."

A software upgrade to be sent up to the rovers in a month or so will provide even greater autonomy. Mission controllers will then be able to tell the rovers simply, "Go to that rock and put your instrument arm down on it." Currently, arm deployment takes an extra day, because controllers must first see the rover's final position before issuing the command to lower the arm.


Another part of the upgrade, adapted from the E.O.-1 software, will enable the rovers to perform a first cut of data, determining what is useful. The rovers have been taking photographs to search for clouds and mini-tornadoes known as dust devils. But through much of the Martian year, clouds and dust devils are rare, so most of the photos show nothing, and beaming them all to Earth is a waste of time.

"You're restricted in the number of images you can return," said Rebecca Castaño of the Jet Propulsion Laboratory.

With the new software, the rovers will analyze the photos and send back only those that appear to contain what the scientists are looking for. That will allow wider and more frequent searches. Dr. Castaño said tests showed that the software was correct in identifying clouds 93 percent of the time.

More important, the software rarely failed at finding a cloud when one was there. Rather, the software sometimes saw a cloud when there was none — a mistake that a scientist on the ground can easily correct.

Still, these spacecraft seem far from intelligent. E.O.-1's scheduling programs are not that different from the software that Wal-Mart uses to manage inventory, and the rovers' driving autonomy does not give them the ability to recognize unanticipated traps. In April last year, the Opportunity blithely drove into a sand dune, and took five weeks to back out.

"None of the A.I. systems are as smart as a 2-year-old," said Cynthia Y. Cheung, a scientist at NASA's Goddard Space Flight Center in Greenbelt, Md. "We want this system to be able to learn and react to the environment."

Dr. Cheung is a member of a team led by Steven Curtis designing a more ambitious rover. It does not have wheels. Instead, it looks like a shape-changing jungle gym, with trusses that lengthen and shorten. A simple prototype has been built.

Computer animations illustrate its possibilities. Across flat terrain, it would roll like tumbleweed.

It could pull itself, almost catlike, onto rocks or flatten itself and slither through holes. (The animation can be viewed at nytimes.com/science.)

To achieve those abilities, the machine would need sensors to observe its surroundings and then use the best mode of locomotion. While some safety rules might be explicitly programmed — the equivalent of telling a child, "Do not cross a busy road" — the scientists also will put in programming that allows the robot to learn its behavior through trial and error.

"You'd essentially set up a playground where the robot can perform these simple behaviors," said Michael Rilee, another member of the Goddard team. "It's a lot like what children do when they're very small, and they're just learning to move around."

Dr. Curtis said he thought the technology could be ready for a rover to explore the rockier regions of the Moon in a few years.

How much new technology will be built and used is an unanswered question. "This is a completely different way of doing business," Dr. Chien said. "People are risk averse, for good reasons."

Interplanetary space probes cost hundreds of millions of dollars each, and a mistake in the programming can lose the craft. An attempt last year to get a small craft named DART, Demonstration of Autonomous Rendezvous Technology, to dock autonomously with another satellite resulted instead in DART's collision with the target.

Scientists are also wary about letting a spacecraft throw away information before humans get to sift through it. They often joke about a rover obliviously driving past a dinosaur bone lying on the Martian surface because it had been programmed to recognize only rocks.

But autonomy will also allow them to do more at lower cost.

For some possible missions, like the Titan blimp, "This sort of thing is going to be key for making the most of the mission," said Ralph D. Lorenz, a planetary scientist at the University of Arizona.

"In an ideal world," he added, "you'd downlink every picture, and scientists would have all of the time in the world to look over them, but you don't have that luxury."

Because of the limited communications with Earth, Dr. Lorenz said the craft might summarize its findings — that it has been flying over sand dunes, for example — and send back only a few representative photos instead of images of the entire landscape.

"To get the best science, you want to send down only the data with the richest science concentration," Dr. Lorenz said.

And for that, Dr. Lorenz is willing to let the spacecraft do the choosing.

    Intelligent Beings in Space!, NYT, 30.5.2006, http://www.nytimes.com/2006/05/30/science/space/30rock.html

States again ponder space travel business

Updated 5/14/2006 11:43 AM ET
By Alicia Chang, The Associated Press
USA Today

LOS ANGELES — The promise of blasting thrill-seeking tourists into space is fueling an unprecedented rush to build snazzy commercial spaceports.
The Federal Aviation Administration is reviewing proposals from New Mexico, Oklahoma and Texas to be gateways for private space travel. Depending on how environmental reviews and other requirements go, approval could come as early as this year and the sites could begin preparing to ferry space tourists soon afterward.

The current spaceport boom recalls the mid-1990s, when the first spaceport fad generated hype but no real construction. Finally, technology may have caught up with starry-eyed plans.

Aerospace designer Burt Rutan, who is building a commercial spaceship fleet for British space tourism operator Virgin Galactic, recently expressed his amazement at the flurry of proposals.

"It's almost humorous to watch the worldwide battle of the spaceports," Rutan mused earlier this month at the International Space Development Conference.

For decades, spaceports have been used mostly by NASA and the Pentagon to rocket astronauts and satellites into orbit.

Traditional launch ranges are often spartan mixes of lonely launch pad towers, concrete runways and aircraft hangars. Many are located in remote coastal areas — Florida's Cape Canaveral being the best known — so that debris won't hit populated areas.

The current spaceport boom promises futuristic complexes that evoke the Jetsons. But cashing in requires a gamble.

None of the private rockets under development has been test-flown. And even once the FAA licenses any vehicles, the infant industry initially won't boast multiple daily flights — at $100,000 to $250,000 a head, the market is decidedly limited.

For states that invest early, however, the long-term economic benefits could be substantial. A recent study commissioned by New Mexico predicted that its proposed hub could net $750 million in revenue and up to 5,800 new jobs by 2020. States with spaceports anchored by a reliable spaceliner and designed like a galactic Disneyland also could be a magnet for high-skill, high-wage labor and sprout cottage industries.

The rush to build commercial space hubs is spurred by entrepreneurs who want to send rich passengers into suborbital space — a region about 60 miles above Earth. Several will build their rockets this summer with tentative plans to fly as early as next year pending regulatory approval.

New Mexico, which inked a deal with Virgin Galactic last year to construct a $225 million spaceport on 27 square miles of desert, is expected to select a winning architectural design from six entries on June 2.

While details of the spaceport designs are secret until a winner is chosen, tentative plans call for a complex built mostly underground. The facility, which would be funded by a mix of federal, state and local money, could open in late 2009. Virgin would have a 20-year lease on the facility.

Until then, Virgin Galactic, founded by British mogul Richard Branson, plans to fly the first passengers from California's Mojave Airport, where the Rutan-designed SpaceShipOne became the first privately manned rocketship to reach space in 2004.

Outside the seven government-run launch ranges, the Mojave hosts one of five non-federal licensed spaceports in the U.S. that serve both commercial and government interests. The non-federal ports are either state or privately operated.

Another space tourism operator, Rocketplane Kistler, wants to launch from the Oklahoma Spaceport. Still pending FAA approval, that spaceport sits on the site of the former Clinton-Sherman Airpark in Burns Flats and boasts one of the world's longest runways.

The Oklahoma Spaceport has passed all its requirements and is expected to win an FAA license over the next several weeks, said Bill Khourie, executive director of the Oklahoma Space Industry Development Authority.

Mission control is being upgraded and there are plans for VIP lounges and other amenities.

"We ultimately plan on building more sexy facilities," said Charles Lauer, vice president of business development at Rocketplane Kistler.

The FAA also is considering two proposed spaceports in Texas, including a private spaceport on 165,000 acres of desolate ranch land about 120 miles east of El Paso bought by Amazon.com founder Jeff Bezos. Bezos had said his space tourism firm, Blue Origin, would first build basic structures, then begin flight tests in six to seven years.

To gain a spaceport license, a facility must pass an environmental review and prove that its location won't harm surrounding communities or the public, said Patricia Grace Smith, associate administrator of the FAA Office of Commercial Space Transportation. Spaceports and launch vehicles are licensed separately.

The 1990s saw a different spaceport race.

Back then, New Mexico, Oklahoma and Texas were among more than a dozen states that squandered thousands of dollars trying to woo a much-hyped experimental spacecraft program.

VentureStar, a wedge-shaped spaceship by NASA and Lockheed Martin, was supposed to replace the space shuttle. But the program was plagued by engineering problems and was scrubbed in 2001.

To avoid another bust, commercial space hubs must find creative ways to supplement tourists' weightless experience by adding attractions such as theme parks, hotels and restaurants, said Derek Webber, director of Spaceport Associates, a Maryland-based consulting firm.

"You've got to do your homework," he said, "because not all states will succeed."

    States again ponder space travel business, UT, 14.5.2006, http://www.usatoday.com/tech/science/space/2006-05-14-spaceports_x.htm

Making money on the moon seen key to exploration

Fri Apr 28, 2006 8:24 PM ET
Reuters
By Deborah Zabarenko

WASHINGTON (Reuters) - Making money on the moon is an essential part of the U.S. plan for space exploration, NASA officials said on Friday after a four-day strategy workshop with international space officials and scientists.

Billed as the first meeting to determine what explorers would do if they return to the lunar surface after more than three decades, the gathering drew some 180 participants from more than a dozen countries, including China, Russia, Japan and the nations of the European Space Agency.

Shana Dale, NASA's deputy administrator, said one clear goal was to do business.

"The teams recognize the critical importance of space commerce -- having real companies going to the moon and making money," Dale said at a telephone news conference. "The government needs to be a trailblazer and enabler (with) a desire to see commerce take off."

Other essentials for a global space strategy include public involvement and participation by international partners, Dale said.

The strategy workshop was the first of several scheduled for this year that aim to set out specific goals for future space missions to the moon and Mars, as described by President George W. Bush in 2004 in a sweeping "Vision for Space Exploration."

Delivered less than a year after the fatal 2003 shuttle Columbia accident, Bush called for a human return to the moon by 2020 and eventually a human flight to Mars.

Since the Columbia disaster, in which seven astronauts died, only one space shuttle has flown, and the shuttle fleet is to be retired by 2010.

A new Crew Exploration Vehicle meant to return humans to the lunar surface is still on the drawing board, and may not be ready until 2012 or later.

The last time humans went to the moon was aboard NASA's Apollo 17 in 1972. Since then, China has begun its own human space program and also sent representatives to this meeting, though Dale said they apparently did not attend the smaller working sessions.

Aside from the central issues of commerce, international cooperation and public engagement, the working groups also noted the need for lunar law early in the process.

David Beatty of NASA's Jet Propulsion Laboratory said an international legal framework would be helpful in the area of property rights, interoperability standards and making hardware from various countries work together.

Such laws could govern more prosaic issues as well, Laurie Lesin of NASA's Goddard Space Flight Center said.

"At our group, we did mention once briefly how we're going to decide which side of the road we drive on, on the moon," Lesin said with a laugh.

    Making money on the moon seen key to exploration, NYT, 28.4.2006, http://today.reuters.com/news/articlenews.aspx?type=topNews&storyid=2006-04-29T002440Z_01_N28102069_RTRUKOC_0_US-SPACE-EXPLORATION.xml

NASA launches climate satellites

Fri Apr 28, 2006 9:51 AM ET
Reuters
By Irene Klotz

CAPE CANAVERAL, Florida (Reuters) - NASA on Friday launched two research satellites to help scientists refine computer models that forecast the weather and chart global climate change.

CloudSat and CALIPSO blasted off aboard an unmanned Delta rocket from Vandenberg Air Force Base in California at 6:02 a.m. EDT (1002 GMT) after a week of delays for weather and technical issues. The Boeing-built booster originally had been slated to fly last year, but a machinists' strike forced several months of delays.

CloudSat has powerful radar instruments to peer deep into the structure of clouds and map their water content. Although only about 1 percent of Earth's water is held in clouds, it plays a crucial role in the planet's weather, scientists working on the mission said.

"CloudSat will answer basic questions about how rain and snow are produced by clouds, how rain and snow are distributed worldwide, and how clouds affect the Earth's climate," principal investigator Graeme Stephens of Colorado State University said.

Using instruments 1,000 times more powerful than common meteorology radar CloudSat was designed to render three-dimensional maps of clouds that will identify the location and form of water molecules.

Complementary and virtually simultaneous studies by sister probe CALIPSO will pinpoint aerosol particles and track how they interact with clouds and move through the atmosphere. CALIPSO is an acronym for Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations.

Aerosols are formed by natural phenomena like forest fires and human activity such as driving cars. Aerosols are considered a key factor in understanding why the planet is growing warmer and if anything can be done to stem or reverse the change.

Computer models predict average surface temperatures on Earth will increase between 3.5 degrees Celsius and 9 degrees F over the next 100 years.

The uncertainty stems from the role clouds play in moderating heat. Aerosols in the clouds can either cool the planet by reflecting solar energy back into space, or increase temperatures by trapping heat in the atmosphere.

"We need to understand the aerosol effect on climate because it counteracts the effects of greenhouse gases," said CALIPSO principal investigator David Winker of NASA's Langley Research Center in Hampton, Virginia.

    NASA launches climate satellites, NYT, 28.4.2006, http://today.reuters.com/news/articlenews.aspx?type=scienceNews&storyid=2006-04-28T135118Z_01_B193600_RTRUKOC_0_US-SPACE-SATELLITES.xml

Griffin: 2011 earliest for new spaceship

Posted 4/25/2006 8:54 PM ET
By Suzanne Gamboa, The Associated Press
USA Today

WASHINGTON — A new spaceship could be ready to replace the nation's aging shuttle fleet by 2011 — three years ahead of schedule — if lawmakers added money to NASA's proposed budget, the head of the space agency told a congressional panel on Tuesday.

NASA Administrator Michael Griffin said that date is the earliest the new spaceship, or crew exploration vehicle, could be developed no matter how much money the agency received.

Currently, the target date for building a new vehicle is 2014.

With his pitch to Congress, Griffin underscored a point he has made previously about completing the spaceship on a faster time frame.

Pressed by Sen. Bill Nelson, D-Fla., Griffin acknowledged an additional $1 billion could accelerate the program's completion.

The shuttle is to be retired in 2010, and lawmakers are concerned about when a replacement will be ready.

"If money were not an issue — going back to Apollo kinds of days — then I think it would be no technical problem to have an operational system available in five to six years," Griffin said after testifying before a subcommittee that oversees NASA spending.

President Bush's budget calls for a 3.2% increase in NASA spending over last year. The House and Senate have authorized an additional $1.1 billion, but that is only a guide. The money must be appropriated by both chambers.

A Senate appropriations subcommittee was to scheduled to meet Wednesday to consider the proposed increase.

NASA will be shelving its three aging space shuttles in four years. The next generation of spaceships is supposed to carry astronauts to the moon by 2018 and eventually to Mars.

The so-called "flight gap" between the shuttles and their replacement could affect research and American space competitiveness, lawmakers said.

"We don't want a hiatus because we think that puts us in a security risk position," said Sen. Kay Bailey Hutchison, R-Texas.

    Griffin: 2011 earliest for new spaceship, UT, 25.4.2006, http://www.usatoday.com/tech/science/space/2006-04-25-nasa-shuttle-replacement_x.htm

American, Russian, Brazilian return from international space station

Updated 4/8/2006 10:04 PM
USA Today

ARKALYK, Kazakhstan (AP) — A capsule carrying Brazil's first astronaut, along with a Russian and an American, landed safely in the freezing Kazakh steppe early Sunday after separating from the international space station and hurtling through the Earth's atmosphere.
American astronaut Bill McArthur, Russian cosmonaut Valery Tokarev and Brazilian Marcos Pontes touched down on target and on schedule.

The TMA-7 landed on its bottom about 30 miles northeast of Arkalyk after what Mission Control officials called a flawless flight. Officials at Russia's Mission Control in Korolyov, outside Moscow, reported that the capsule had been in radio contact for much of the bone-jarring, 3{-hour journey and that all three crewmembers were feeling well.

Ground crews reached the capsule in northern Kazakhstan, where temperatures hovered around 13 degrees below zero, within minutes of the landing. McArthur, shown on a Mission Control screen as he was still strapped inside the capsule, looked dazed after the 250-mile trip from the space station.

Pontes, seated in a chair outside the capsule, grinned and gave a thumbs-up as his bulky spacesuit was removed. He was handed a Brazilian flag and a Panama hat that was pulled out of the capsule — apparently one that he had carried to the space station in tribute to the Brazilian inventor and aviator Alberto Santos Dumont, to whom Pontes had dedicated his flight.

The three travelers were given hot tea and wrapped in blankets before being whisked into a medical tent.

McArthur and Tokarev had spent more than six months on the space station. They were replaced by Russian commander Pavel Vinogradov and U.S. flight engineer Jeff Williams, who arrived at the station together with Pontes on April 1. Pontes had traveled to the station for a weeklong stint.

Speaking at Russian Mission Control, the head of Brazil's space program, Raimundo Mussi, said Ponte would receive "a big hug from all Brazilians" upon his return home.

The American space program has depended on the Russians for cargo and astronaut delivery since the February 2003 Columbia disaster grounded the shuttle fleet. The shuttle Discovery visited the station last July but problems with the external fuel tank's foam insulation have cast doubt on when shuttles might return to flying.

    American, Russian, Brazilian return from international space station, UT, 8.4.2006, http://www.usatoday.com/tech/science/space/2006-04-08-space-station-return_x.htm

Nasa to put man on far side of moon

March 19, 2006
The Sunday Times 
Jonathan Leake , Science Editor

NASA, the American space agency, has unveiled plans for one of the largest rockets ever built to take a manned mission to the far side of the moon.

It will ferry a mother ship and lunar lander into Earth orbit to link up with a smaller rocket carrying the crew. Once united they will head for the moon where the larger ship will remain in orbit after launching the lunar lander and crew.

The design emerged during a space science conference in Houston, Texas, last week. The plan is part of Nasa’s “Return to the Moon” programme set in motion by President George W Bush two years ago.

Under the project, up to four astronauts at a time will land on the far side of the moon to collect rock samples and carry out research, including looking for water that might one day support a lunar base.

The scale of the missions is much larger than the earlier Apollo programme, which is why Nasa will need two separate rockets to take the mother ship and crew into space.

Some missions will also see manned spacecraft landing in unexplored areas such as the lunar mountains and on the moon’s south and north poles.

John Connolly, manager of Nasa’s lunar lander project, said the system was designed to carry crews to almost every part of the moon’s surface.

“The samples they collect and the research they carry out will help solve many mysteries about the origins and composition of the moon and its suitability as a base,” he said.

The Apollo programme carried out six lunar landings between 1969 and 1972. The feat was a triumph, but the technical limitations of the Apollo craft, plus ignorance of lunar terrain, meant all six missions had to be sent to the moon’s plains.

These regions, all on the near side of the moon, were the only areas known to be flat enough for a safe landing. This has frustrated scientists because the samples collected by the six missions are all similar. They are also thought to be younger than lunar mountain rocks.

The far side — so called because it always faces away from the Earth — was first photographed in 1959 by a Russian probe. In 1968 the astronauts of Apollo 8 became the first to view it directly.

The evidence gathered by such missions was enough to deter any attempt to land because most of the far side appeared to be covered in large craters. Additionally, any craft landing there would be cut off from radio contact with Earth.

Connolly believes, however, that Nasa will be able to overcome such problems by sending a series of robotic probes ahead of the manned missions.

The first of these, the Lunar Reconnaissance Orbiter, to be launched in 2008, will map the moon’s surface in detail.

Cameras will photograph the surface, backed by a laser altimeter to create a three-dimensional relief map from which Nasa can identify landing sites.

Then, from 2010, a series of “companion lander” missions will carry out test landings on selected sites to see if they are worth a visit by humans.

The final element will be a system of communications satellites, dubbed the “lunar internet”, so astronauts will be able to relay signals to Earth from any part of the moon.

Connolly said the first humans could arrive as early as 2015, although 2018 was more likely. The agency would then aim to send two crews to the moon each year for up to five years. The programme will cost around £56 billion and may also be used to test technology for any future mission to Mars.

Some have questioned whether the programme will produce enough good science to justify the costs.

Manuel Grande, head of the planetary science group at the Rutherford Appleton Laboratory in Oxfordshire, dismissed such fears. “Finding out more about the moon will help us understand where the Earth and moon came from,” he said. “There do not have to be good scientific reasons . . . It’s like going up Everest; we want to go to places like the moon and Mars just because they are there.”

    Nasa to put man on far side of moon, ST, 19.3.2006, http://www.timesonline.co.uk/article/0,,2089-2092495,00.html

A Bold Plan to Go Where Men Have Gone Before

February 5, 2006
The New York Times
By LESLIE WAYNE
EL SEGUNDO, Calif.

ASK Elon Musk what he wants to do with his life — after having amassed a $300 million fortune from the Internet — and the answer is surprising. At 34, he says he is too young to retire. Philanthropy is a bit staid. Starting another Web-based venture is hardly a challenge, not for a man who bought the idea for PayPal, built it up and then sold it to eBay for $1.5 billion.

In seeking a new direction in life that would be as ambitious as his dreams, Mr. Musk has picked a doozy: cheap and reliable access to space.

Making money from space is a road that several other self-made millionaires have traveled, from a Texas banker named Andrew Beal to one of Microsoft's co-founders, Paul G. Allen. There have been enough of them to warrant a mocking nickname: "thrillionaires." And so far their efforts have either ended in failure or have been just ventures in "space tourism" that brought test pilots to the fringe of space.

Mr. Musk wants more, and he has put $100 million of his fortune on the line to try to get it. His goal is to make a business out of inexpensively launching satellites into orbit. Inexpensive, of course, is a relative term, in a business where launchings of private commercial weather, telecommunications and other payloads start at $30 million and go up to $85 million or more.

Through his company, Space Explorations Technology, or SpaceX, Mr. Musk wants to send things to space for one-third of the going rate or less — even bringing down the price to $7 million for small payloads to low Earth orbit — with a series of simple rockets of his own design. His goal is to build a Volkswagen of the cosmos, a bare-bones and dirt-cheap rocket that will go into space and return, to be used again and again. Commercial launchings currently cost $5,000 to $10,000 per pound of payload; Mr. Musk says his simple rockets could do it for $1,000 a pound.

His first rocket, the Falcon 1, is a two-stage, liquid fuel design that is scheduled to lift off on Wednesday from a United States Army facility on Omelek Island in the Kwajalein Atoll, part of the Marshall Islands. On board will be a 43-pound satellite, the FalconSAT-2, which was designed by Air Force Academy cadets to study the ionosphere.

THE launching has been postponed twice for technical reasons, but if it succeeds, it will move SpaceX closer to filling some of the $200 million in firm launching orders already placed by the Pentagon, foreign governments and private companies.

Less clear is whether a success will also silence the many skeptics who have seen wealthy space dreamers fail in the past.

"This is an enormously difficult business to make money in," said John E. Pike, a space policy analyst at GlobalSecurity.Org, a nonprofit group in Alexandria, Va., that analyzes national security issues. "The best way to make a small fortune in space is to start out with a large one. New rocket science has a high mortality rate, and we don't know what he's got his hands on until he's flown it a half-dozen times."

Part dreamer and part realist, Mr. Musk says he was drawn to the project not only because he has long been fascinated by space — he has a degree in physics from the University of Pennsylvania — but also because he sees a market opportunity in America's declining share of the world's satellite-launching business.

In the commercial market, the United States' two big rocket giants, Lockheed Martin and Boeing, have been priced out by lower-cost competitors from Russia, Ukraine and France. Lockheed's Atlas 5 had only one commercial order in 2005, compared with 22 in 1998. Boeing has withdrawn its Delta 4 rocket from the commercial market and relies exclusively on business from the United States government.

At stake is a market that was worth $4 billion last year, when governments and businesses paid for 55 launchings, according to the Federal Aviation Administration. Of those, 18 were commercial, with a value of $1 billion.

American companies compete for commercial orders only by teaming with foreign partners — often former cold-war foes. Lockheed has teamed up with Khrunichev State Research of Russia to form International Launch Services, which mainly uses Russia's Proton rockets. Boeing has joined with several nations to form a consortium called the Sea Launch; it uses the Ukrainian Zenit 3SL to put up commercial payloads.

Mr. Musk says he wants to develop an all-American option that will be price-competitive and break the duopoly of Lockheed and Boeing on contracts with the federal government. Ultimately, he wants to send people into space, to the moon and beyond.

"We have to do something dramatic to reduce the cost of getting to space," said Mr. Musk in an interview in his cubicle at SpaceX's offices here. "If we can get the cost low, we can extend life to another planet.

"I want to help make humanity a space-faring civilization," he said with disarming — or alarming — candor.

SpaceX's first effort, the Falcon 1, will not put anyone on the moon. It is designed to send small satellites — typically communications and scientific payloads weighing less than 1,000 pounds — into low orbit, which is up to 300 miles above the Earth. The two-stage Falcon 1 is designed to be mostly recyclable, with part of it falling into the ocean to be picked up and used again.

The Falcon 1 will charge $6.9 million a launching. It is intended to go head to head with the Pegasus rocket made by the Orbital Sciences Corporation of Dulles, Va., which charges $25 million to $30 million for the same launching, as well as rockets from such newcomers as India and Israel.

Next up are the Falcon 5, the same rocket with five engines, and the Falcon 9, with nine engines. The Falcon 9 would bring SpaceX into direct competition with Boeing's Delta 4 and Lockheed's Atlas 5 in the so-called heavy-lift market, in which the United States government is the main customer. A Falcon 5 plans to launch 8,000-pound payloads for $18 million, a third of the price of competitors. The Falcon 9, which will put 10 tons of payload as far as 22,000 miles into the sky, will cost $27 million per launching. The same launching by Lockheed or Boeing would be about $70 million to $80 million.

Expecting that it can compete in this market, SpaceX has sued Boeing and Lockheed in federal court in California, seeking to prevent them from combining their rocket units in a joint venture called the United Launch Alliance, which would have a lock on $32 billion in Air Force launchings through 2011.

"SpaceX has the potential of saving the U.S. government $1 billion a year," Mr. Musk said. "We are opposed to creating an entrenched monopoly with no realistic means for anyone to compete."

A native of Pretoria, South Africa, Mr. Musk moved by himself at age 17 to Canada, where he briefly attended Queens University in Kingston, Ontario. He later transferred to the University of Pennsylvania, where he received two undergraduate degrees — one in theoretical physics and the other in business from the Wharton School. He enrolled in the graduate physics program at Stanford in 1995 but dropped out within days to become an Internet entrepreneur.

His first big hit was Zip2, a Web-based ad company that he sold in 1999 for $307 million. (The New York Times Company was an investor in Zip2.) He moved on to another Web venture, involving electronic payments over the Internet — even though, as skeptics noted, he lacked experience in banking. That business ultimately became PayPal, which eBay bought for $1.5 billion in 2002.

By the time Mr. Musk was 30, he had amassed $300 million and was pondering his future.

His first thoughts were of philanthropy — and of space. He came up with the idea of "Mars Oasis," an effort to send a small greenhouse to Mars to gather scientific information and create excitement about space travel — or so Mr. Musk thought. His idea was quickly derailed by the extraordinary cost of getting to space, but that led him to wonder why technology had not brought down the cost of space exploration or led to more of it. And that led him to found SpaceX in 2002.

Every day since then, Mr. Musk has driven to a gritty industrial zone, where he puts in long days at SpaceX. Still, he allows himself a few perks of the newly rich: a big house in Los Angeles, a $1 million McLaren F1 sports car, and a Dassault Falcon 900 business jet, which he sometimes uses to ferry his staff to the Marshall Islands.

In some ways, SpaceX is a throwback to his dot-com days. Many of the 160 employees, including former engineers from Boeing and other aerospace companies, are on a first-name basis with him. One building houses a Ping-Pong table; another has a Segway. All employees — who call themselves "SpaceExers" — have received stock options that could make them millionaires someday. In one spot, a blue tarp covers a small piece of a rocket that Mr. Musk casually described as a "top secret" project and joked about putting a sign on it saying so. Indeed, it was a part for a launching scheduled by the Pentagon, which already has $100 million of SpaceX business lined up. He has another $100 million in launchings from the government of Malaysia, the Swedish Space Corporation and several American companies, including Bigelow Aerospace, which is planning to build a private space station.

TheFalcon 1 flight this week is for the Air Force and the Defense Advanced Research Projects Agency, the Pentagon's research and development arm. "DARPA is excited about the launch," said Steven Walker, the DARPA manager for the Falcon program. "A successful launch demonstration will change the way we do space launch in this country."

WHAT sets SpaceX apart from other rocket makers is that the boldness of its ambition is matched by the modesty of its design. To meet his goal of a cheap and reliable rocket, Mr. Musk is producing a basic design, with fewer opportunities for systems to fail — even if it means some technical compromises with performance. "SpaceX is optimizing for simplicity rather than performance, and that's what sets it apart from the others," said Jeffrey Foust, an analyst at the Futron Corporation, an aerospace consulting firm. "When you have a limited number of things that could fail, you can increase a rocket's reliability."

Where most other rockets have multiple stages and multiple engines, the Falcon will have just two stages, each with one engine. Most of SpaceX's stages are designed to be reusable. Although fishing small used rockets out of a vast ocean can be difficult, Mr. Musk says that it is cheaper than building a new one every time.

"Throwing away multimillion-dollar rocket stages every flight," he said, "makes no more sense than chucking away a 747 after every flight."

Instead of buying engines from established suppliers, Mr. Musk has designed his own and built them in-house. The bigger first-stage engine, called the Merlin, is a model of 1960's technology, a simple "pintle" engine that has only one fuel injector rather than the costly showerhead of injectors used in most rockets.

"The Merlin is much more analogous to a truck engine than a sports car engine, which is how all other engines are designed," he said. "Instead of designing it to the bleeding edge of performance and drawing out every last ounce of thrust, we designed Merlin to be easy to build, easy to fix and robust. It can take a beating and still keep going."

AS such remarks make clear, economy is everything at SpaceX. While Boeing and Lockheed typically have more than 200 people in their mission control launching centers, SpaceX will have 23, in part because all the Falcon's manufacturing is done in the factory and nothing is left to be assembled at the launching pad. SpaceX plans some launchings for Vandenberg Air Force Base near Lompoc, Calif., where it will move its mission control, which is now housed in a truck in SpaceX's parking lot.

SpaceX also has a 300-acre test site in McGregor, Tex., that it bought from Mr. Beal, the Texas banker, after he abandoned his private space-launching effort. He closed his company, Beal Aerospace, in 2000, complaining that Boeing, Lockheed and NASA had a lock on launchings and that small entrepreneurs could not compete against these government-subsidized ventures.

Mr. Beal and others have said that the biggest problems for space entrepreneurs are more political than technical. For Mr. Musk, wealth provides some protection — a point that even critics like Mr. Pike of GlobalSecurity.Org concede. "He's got the advantage of deep pockets," Mr. Pike said.

For the moment, Mr. Musk is bankrolling SpaceX alone. But if he can launch Falcon 1, he anticipates getting venture capital money along with more commercial orders.

"There is concern that the United States is losing its competitive edge in commercial space launches," said Mr. Foust, the Futron analyst. "If SpaceX can provide low-cost launches that are reliable, it could turn the tide. He's certainly got the mind-set, the team and the money."

Most other "thrillionaire" ventures revolve around space tourism and suborbital trips, which are less challenging and costly.

Mr. Allen, the billionaire co-founder of Microsoft, provided $25 million to help bankroll SpaceShipOne, which was designed by the aeronautical engineer Burt Rutan and won the $10 million Ansari X Prize in October 2004 for being the first private manned spacecraft to reach suborbital space twice. Mr. Rutan is now designing a bigger version for Virgin Galactic, the space tourism venture of Sir Richard Branson, who founded Virgin Atlantic Airways.

The engines on Mr. Rutan's SpaceShipOne came from SpaceDev, founded by Jim Benson, a computer engineer who had started two software companies. SpaceDev plans to put satellites on Falcon 1 and is developing the "Dream Chaser" to pursue suborbital space tourism.

Jeff Bezos, founder of Amazon.com, has started Blue Origin to develop a three-person suborbital rocket. And John Carmack, developer of the Doom and Quake computer games, has founded Armadillo Aerospace near Dallas and has already put in $1 million to build a suborbital craft.

Space tourism presents no threat to Boeing or Lockheed, but Mr. Musk could. The two companies are quick to dismiss him. "Launching into space is an extremely challenging and complex business," said Dan Beck, a Boeing spokesman, adding: "For SpaceX to be considered a potential competitor they need to have a launch."

Tom Jurkowsky, a spokesman for Lockheed, had a similar view. "SpaceX needs to prove themselves," he said, "and thus far they have been unable to demonstrate that they are a competitor."

Still, some cheer on Mr. Musk.

"I'm particularly happy to see it happen," said Robert Sackheim, chief propulsion engineer at NASA's Marshall Space Flight Center and an early consultant to Mr. Musk. "Their engine design is less than perfect, but it is good enough. I think he is doing all the right things. This can be an incredibly important advance to the country."

    A Bold Plan to Go Where Men Have Gone Before, NYT, 5.2.2006, http://www.nytimes.com/2006/02/05/business/yourmoney/05rocket.html