CNN/AP: Report: Robotic spacecraft crashed into satellite

[DeepModem Mom]

Report: Spacecraft crashed into satellite
Monday, May 15, 2006


The DART spacecraft shut down halfway into its 24-hour mission and failed to complete several tasks.

LOS ANGELES, California (AP) -- A robotic NASA spacecraft designed to rendezvous with an orbiting satellite instead crashed into its target, according to a summary of the investigation released Monday.

Investigators blamed the collision on faulty navigational data that caused the DART spacecraft to believe that it was backing away from its target when it was actually bearing down on it....

***

The 800-pound Demonstration for Autonomous Rendezvous Technology spacecraft was supposed to rendezvous with a defunct Pentagon satellite during a 24-hour period last year....

***

Last month, NASA said it won't release the investigative board's full 70-page report, citing sensitive information protected by International Traffic in Arms Regulations. The summary was prepared by the space agency's exploration systems mission directorate.

Robotic technology plays a critical role in NASA's plan to send humans back to the moon and Mars. The $110 million DART mission was meant to test whether robots can perform some of the tasks astronauts currently must do....

http://www.cnn.com/2006/TECH/space/05/15/spacecraft.mishap.ap/index.html

[PATRICK]

1. NASA wins anti-missile contract

Looks like they now have a better test record than the Star Wars program.

Yippee!

Conversely, should they design robot troops for Iraq their directional systems would soon have lead to a full withdrawal. Send NASA robots to replace the human troops in Iraq!

[unhappycamper]

2. Looks like the satellite killer is now functional.

Another dot.....

[TX-RAT]

7. Kinda what i thought.

[jpak]

15. Bingo

Abrogate ABM Treaty - check

Disembowel Nuclear Nonproliferation Treaty (India nuke deal) - Check

Militarize Space - Check

End Taboo on first use of nucular weapons (Divine Strake and mini-nukes) - Coming soon to an I-ranian unranium enrichment plant near you...

[Atman]

3. Oh, the irony!

They spend a brazillion dollars on Stars Wars, and they can hit targets that are rigged in the first place...but when they DON'T want to, they knock satellites out of the sky. Good work, BushCo!

[DS1]

4. Those > or < motion calculations will always get ya

:eyes:

[Canuckistanian]

6. "R" and "D" are hard for robotic cameras to distinguish!

Cut them some slack!

[PCIntern]

5. ooops, we did it again...

On the one hand, we have 5 fighter jets protecting the entire east coast...as witnessed by our 9/11 response. Lucky NASA wasn't managing them...they'd have all collided out over the Atlantic Ocean where they had them all flying while the hijacked planes were headed towards DC.

[GodlessBiker]

8. ...when it was actually bearing down on it....D'OH!

[Ezlivin]

9. I can feel my mind going, Dave

So it'll be a bit longer before we have HAL running our space missions, huh?

[MaineDem]

10. Maybe this was the plan all along

You know, I'm just saying...

[Psephos]

11. Nothing personal guys, but the level of ignorance on this thread is high

This is a technology *development* program, and the technology involved (autonomous robotics) relies on immensely complex software that must operate without direct human intervention. No one knows exactly how millions of lines of computer code will interact until that code is put to a real life test. (See "Vista, Microsoft" for further insight.)

NASA learned a lot more from the failure in this program than they would have if it had gone perfectly. A LOT more. Technological progress *requires* pushing into failure-prone modes - that's how learning happens.

If test programs don't have frequent failures then they aren't really test programs. Advance happens at the margins, not in the comfortable middle. This is true in personal life as well. RIGHT?

Peace.


edit: typo

[VegasWolf]

12. Not this time, you might think that the directional component of the

algorithms was a fairly important (read critical) component. A negative distance indicator should have been caught.

[formercia]

13. wrong metrics

calculating velocity in furlongs per fortnight....

[VegasWolf]

14. in fathoms per year!!! nt

[MadHound]

17. Lighten up Francis

Geez, we all know the technology involved, the risks involved, and how NASA and others learn from their mistakes. Most people here are just having a little fun with this OK.

[Spinzonner]

18. Failure to test and detect fundamental problems like that

is an obvious management failure that has been characteristic of too many NASA project over the years including the Mars expedition that used incompatible metrics and the loss of two Space Shuttles.

[Psephos]

19. "Obvious" to anyone who's *not* a software or systems engineer

But clearly *not* obvious to the actual software and systems engineers who put this bird up there.

It's easy for those who don't actually do the work to bitch about it; I guess it's just human nature.

The failure, if you RTFA, was anything but a fundamental problem. Bugs are by definition unforeseen, and their effects cascade in unpredictable ways, as happened in the current example. When developing aerospace systems, engineers must hunt down thousands of these. Catching 99% of them isn't good enough...you have to get 99.999999% of them ("six-sigma" quality).

I suggest reading a book or two by Henry Petroski if you're interested in how engineers really work, and how much of the world we truly owe to them.

Here's an excellent one to start with:

Success through Failure - The Paradox of Design
http://www.amazon.com/gp/product/0691122253/sr=8-1/qid=1147808227/ref=pd_bbs_1/102-4247228-4282534?%5Fencoding=UTF8

Peace.

[Spinzonner]

20. But I am a software engineer

and am very familiar with he consequences of truncated test and QA being applied to projects all along the process.

You might look at the reports on the problems with the Mars mission that suffered from the metrics problem and how inadequate management and controls contributed to the error not being detected at many points along the way.

And you might watch your assumptions about people just because they don't agree with your POV.

[VegasWolf]

21. And I a software engineer and research engineer who has worked on many

large scale project. This component should have been heavily tested. Jeez, even a rudimentary software test should have caught this.

[NoodleyAppendage]

22. You assume that the purpose was not to destroy.

It may well be that the test went off completely as planned, but not as advertised.

J

[VegasWolf]

23. Then that is just what we need, a whole bunch of new debris in the

earth sattelite orbit region. Great!

:sarcasm:

[dysfunctional press]

25. if things had gone perfectly...that would mean they got it right...

and therefore nothing more needed to be learned on that step...
so saying that they learned more from a failure than a success is a little disingenuous.

this kind of ranks up there with the failed mars mission that occurred when one team used feet, and the other used meters...seems like an expensive way to learn about confirming measurements...or maybe i'm just ignernt...

[karlrschneider]

16. Oops. The formula was for Km not miles!

:eyes:

[The Stranger]

24. I love it when the rocket scientists have a flame war on the most abstruse

of topics.

The algorithm software superstitial co-efficient techno-space age nano-particle gravitational vector constants were reversing bi-spatial directional and anti-directional components.

So the rocket crashed.

[Chrisduhfur]

26. lol

:D haha gotta love the internets.....

[daleo]

27. I wonder if it was "accidentally on purpose"

With a bit of "send a message" too.

[Endangered Specie]

28. "Warning: Warp core breach... ALOT SOONER than you think!"

To those of you who think they were 'intentionally' trying to hit that satellite, you should have realized... if they were really trying to hit it, they would have missed by several miles.

[daleo]

29. Well, that does make a kind of sense

In the Bush reverse bizzarro world we are forced to inhabit.