Spacecraft Crashes Into Satellite

Juha-Matti Laurio writes "A robotic NASA spacecraft designed to rendezvous with an orbiting satellite instead crashed into its target. Unbeknownst to engineers at the time, DART's main sensor mistakenly believed it was flying away from the satellite when it was actually moving 5 feet per second toward it, investigators found."

ABC News is never the best choice

[caryw]

Offical NASA writeup available here: http://patriot.net/~cary/slashdot/dart_mishap.html [patriot.net]

Made from original PDF available here: http://www.nasa.gov/pdf/148072main_DART_mishap_ove rview.pdf [nasa.gov]

(I hate PDF's for simple text things like this)

--
NoFluffNews.com - Currently in development but seeking journalists and editors [nofluffnews.com]

Old news

[Anonymous Coward]

This is actually year old happening. News is that Nasa finally made public report about it.

Re:Oddly familiar

[ObsessiveMathsFreak]

I know what you mean - you'd have to teach him calculus before he'd understand.

Not really. Just tell then it's the area under a curve, or the volume under a sheet. Even the most pretentious manager will be able to grasp that.

Re:how is this different from Progress ?

[rtz]

One big difference is that Progress needs guidance beacons on the ISS, DART is supposed to be self-contained, able to dock using only it's own sensors.

Ignorance of engineering practices

[mark_jabroni]

It really bothers me to hear this, as an engineer. I hate listening to the media about stuff like this, because they have absolutely no knowledge of engineering methods, and they don't seem care.

Anyway, on a big scary program, here's how these sorts of problems are spotted :

1. Mid or low-level engineers spot potential problems
2. They then tell engineering leadership that they are worried about a particular problem.
3. Engineering leadership and/or management then (either informally or through a process called "risk managment") decides whether or not the problem should be addressed.

Step #3 is about as important as step #1, because you absolutely cannot fix every problem. There's neither the time, nor the money.

Something else to keep in mind : if I spotted a problem that would surely doom my project, and can't get engineering leadership/management to agree with me, I should share some of the blame.

A programming error caused it

[SammoJ]

From http://patriot.net/~cary/slashdot/dart_mishap.html [patriot.net]:

In DART's case, the MIB determined that the first cause for its premature retirement occurred when the estimated and measured positions differed to such a degree that the software executed a computational "reset." By design, this reset caused DART to discard its estimated position and speed and restart those estimates using measurements from the primary GPS receiver.

Careful examination of the software code revealed that upon reset, the velocity measurement from the primary GPS receiver was introduced back into the software's calculations of the spacecraft's estimated position and speed. If the measured velocity had been sufficiently accurate, the calculations would have converged and resulted in correct navigational solutions. However, DART's primary GPS receiver consistently produced a measured velocity that was offset or "biased" about 0.6 meters per second from what it should have been. This had the unfortunate effect of causing the calculations, which were being performed autonomously, to once again diverge until the difference became unacceptable to the pre-programmed computer logic. Once the limit as to how much the calculations could differ was reached, the software executed another reset. As a result, this cycle of diverging calculations followed by a software reset occurred about once every three minutes throughout the mission. These continual resets caused the incorrect navigational data that prompted excessive thruster firings and the higher than expected fuel usage.

Too many pre-programmed resets to the wrong data. Oops. I guess one small thing can cause a whole mission to fail! If you read the report though the rest of the mission was pretty succesful. You would have though they would build in some sort of fail safe ground control method though when there's millions of dollars worth of equipment at stake?

Re:Oddly familiar

[Brushfireb]

"Management IS OVERHEAD. You can run a company profitably with only workers (it will most likely be inefficient, but it WILL run). If you only have management, you will not survive."

Wrong, And just plain silly. Without management no work gets done, becuase there is no incentive. If everyone was just going to get their paycheck, no matter what they did, then not many would work. You can not run a company without some form of management. Can you teach school without teachers? Can you have a courtroom without judge? Can you have a family without a parent? Management and Leadership have been existent in society since society has existed, and probably before (Alpha Male In Animal Packs, etc).

Good management has a necessary role in any company or organization. Management provides vision, linkages between department (IE figure out compromises between IT and Sales), and makes the organization possible. A lopsided organization either way is incredibly bad (Too Much Management, or Not Enough).

In the case of the topic at hand, management is still not overhead. BUt they are replaceable, and should be. Besides, all those who are at a company and who dont like or appreciate the management role, should just leave. Go start your own company, hire employees, and then you will begin to understand the role of management.