SoylentNews is people
SoylentNews
A European Space Agency satellite risks colliding with a piece of space debris about 15 centimeters (a half-foot) long this week, forcing ESA's flight control to plan a rare evasive maneuver.
A piece of an old Russian satellite called Cosmos-375 is forecast to miss Swarm-B, one of ESA's three Swarm satellites that measure Earth's magnetic fields, by just over the length of a football field. But the margin of error for that forecast is around 1,000 meters (3,280 feet or more like three football fields).
ESA has been working with data from the US armed forces' Joint Space Operations Center (JSpOC), located at Vandenberg Air Force base in California, to plan a collision avoidance maneuver that would be uploaded to the satellite Wednesday.
If the satellite is able to alter its orbit as planned, the piece of junk should pass 746 meters (2,448 feet) in front of Swarm-B and 56 meters (184 feet) below it.
Pretty interesting that they are able to track a 15cm piece of debris.
Source:
https://www.cnet.com/news/european-space-agency-orbiter-russian-satellite-space-junk-this-week/
-- submitted from IRC
(Score: 2) by Dunbal on Thursday January 26 2017, @03:26PM
When your margin of error is plus or minus 10 times the value you compute, you are doing math wrong and attributing significance to figures that should not be significant.
(Score: 0) by Anonymous Coward on Thursday January 26 2017, @03:57PM
So if I have a Gaussian curve with a mean of 10, but a standard deviation of 100, how am I supposed to describe that function? Am I no longer allowed to tell you it has a mean of 10 in your maths, or do I have to say it has a mean of zero because that is the digit in the third position?
(Score: 2) by Dunbal on Thursday January 26 2017, @11:42PM
So if I have a Gaussian curve with a mean of 10, but a standard deviation of 100
Standard deviation != margin of error.
(Score: 0) by Anonymous Coward on Friday January 27 2017, @11:17PM
In technical fields it is used that way (at least 90 +/- 7% of the time). Your margin of error is typically expressed in standard deviations (3-sigma, 5-sigma, etc.), so this thing should pass 746 meters +/- 500 meters in front (or whatever the hell the numbers are; the summary is confusing).
(Score: 3, Informative) by FatPhil on Thursday January 26 2017, @04:53PM
Imagine an infinite cylinder going througn space, radius 1000m. That is the path of the junk, with the error bars giving it width.
Select a point 50000m from that cylinder. That's the satelite. Do you not agree that it's perfectly valid, perfectly meaningful, and perfectly useful to say in conclusion to this setup that "the junk is expected to have a closest approach of between 49000m and 51000m"? Or equally "a closest approach of 50000m, with an error bar of 1000m"?
Now instead select a different point inside the cylinder, in fact pretty much right in the very centre of its cross-section. That's the satelite and it looks doomed. Do you not agree that it's perfectly valid, perfectly meaningful, and perfectly useful to say in conclusion to this alternative setup that "the junk is expected to have a closest approach of between 0m and 1000m"? Or equally "a closest approach of 0m, with an error bar of 1000m"?
If one is fine, but the other isn't, then you are going to need to describe very precisely what you think the difference between the two is.
Yet the latter has a margin of error which is unimaginably higher than the value computed as the expected distance from the junk's path.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves