Quote from: Robotbeat on 08/30/2016 02:06 amQuote from: MattMason on 08/30/2016 12:39 amSome thoughts:1) No nation or private group is ever going to commit to a "starship" or probe until more is known about its destination.This statement is trivially true. There is serious scientific hay to be made by JWST and the 39m E-ELT, instruments already under construction, let alone ones that will be built in the future.Quote2) As someone said, a space probe won't work. Unless we can create true artificial intelligence for it to operate itselfLike what? think of a probe as just a sensor. It just needs to receive information and broadcast it. Quoteit would be increasingly impossible to control,You pre-program it with instructions, obviously.Quote activate (you're not going to let it run its power and fuel out over years)And why not? Voyager 1 and 2 look like they'll probably last for half a century, which is longer than Breakthrough Starshot will need (for instance). Quoteor even receive the data it finds because the distance needed to send and receive signals aren't going faster than light (and let's leave the SF out of it until someone invents something truly FTL).Why does the probe need to receive commands? Just program it to send the right information. It doesn't need "AI" except in the trivial sense that is already used today on Mars.There's never been a probe that works autonomously. Probes can't anticipate the unexpected or fix issues themselves. From the Dawn probe, JAXA's Akatsuki to MER Spirit, human intervention is eventually required. Now multiply that matter when an issue is reported days, if not months later, and a command is sent to correct an issue when the probe might be pointed the wrong way or dead by the time the command gets there.Durability of probes isn't my problem. JPL continually makes probes that outlast their mission design by years to decades. Now am I complaining about the versatility of many probes to fix many of its problems on its own. The problem is simply communication to fix what the probes cannot.Voyager 1 and 2 have completed their missions.Oh, we should also consider what power source these things would use. RTGs have a good life, but even they might have limits if a mission takes decades, and the scarcity of plutonium is one reason why Juno is using solar panels.You're still talking years to send or receive anything. That can be managed except for some mission-critical issues that have occurred on missions before. Good planning and innovation might reduce those issues.
Quote from: MattMason on 08/30/2016 12:39 amSome thoughts:1) No nation or private group is ever going to commit to a "starship" or probe until more is known about its destination.This statement is trivially true. There is serious scientific hay to be made by JWST and the 39m E-ELT, instruments already under construction, let alone ones that will be built in the future.Quote2) As someone said, a space probe won't work. Unless we can create true artificial intelligence for it to operate itselfLike what? think of a probe as just a sensor. It just needs to receive information and broadcast it. Quoteit would be increasingly impossible to control,You pre-program it with instructions, obviously.Quote activate (you're not going to let it run its power and fuel out over years)And why not? Voyager 1 and 2 look like they'll probably last for half a century, which is longer than Breakthrough Starshot will need (for instance). Quoteor even receive the data it finds because the distance needed to send and receive signals aren't going faster than light (and let's leave the SF out of it until someone invents something truly FTL).Why does the probe need to receive commands? Just program it to send the right information. It doesn't need "AI" except in the trivial sense that is already used today on Mars.
Some thoughts:1) No nation or private group is ever going to commit to a "starship" or probe until more is known about its destination.
2) As someone said, a space probe won't work. Unless we can create true artificial intelligence for it to operate itself
it would be increasingly impossible to control,
activate (you're not going to let it run its power and fuel out over years)
or even receive the data it finds because the distance needed to send and receive signals aren't going faster than light (and let's leave the SF out of it until someone invents something truly FTL).
It seems to me that a probe to Proxima b is not a good idea.Microprobe flybys (e.g. lightsail-propelled stuff) would have hard time collecting useful information: they can snap some pics at best, but how they would transmit the data back? Not trivial, you'd need a HUGE receiving antenna in Sol system.
Macroprobes we don't quite have propulsion for. Need to develop some fusion drive.Developing a VERY powerful space telescope (interferometry?) can give you about the same information, sooner, with less R&D and mission risk (if something does not work out, you aren't at square one. You can debug it).And also the 'scope can look and many, many other targets too, whereas probes capable of reaching Proxima will become even less tenable for targets farther away.
These are the easiest problems to solve in the whole mission.
All the probe needs to do is collect data and relay it to Earth.
Falcon 9 is totally autonomous once launched. Sputnik 1 was autonomous.
New Horizons, when it entered flyby phase, was entirely autonomous for the same reason as a probe to Proxima b would be. It was given all its instructions of where to look beforehand.
You're making a mountain of a molehill. The real challenge is getting there.
Quote from: Robotbeat on 08/31/2016 12:14 amNew Horizons, when it entered flyby phase, was entirely autonomous for the same reason as a probe to Proxima b would be. It was given all its instructions of where to look beforehand.No, it was preprogrammed, not autonomous....
...QuoteYou're making a mountain of a molehill. The real challenge is getting there.Disagree. You are completely failing to grasp the amount of analysis and effort that goes into real spacecraft operations, and how much would have to be done by the spacecraft given a multi-year time lag and limited knowledge of the targets.
Quote from: hop on 08/31/2016 02:08 amNo, it was preprogrammed, not autonomous....Sure, that works.
No, it was preprogrammed, not autonomous....
Quote from: Robotbeat on 08/31/2016 02:13 amQuote from: hop on 08/31/2016 02:08 amNo, it was preprogrammed, not autonomous....Sure, that works.Except it doesn't work for an interstellar mission. That's the whole point. You cannot apply anything like the approach used by New Horizons or any other real space mission to date.And yes, the disagree was only intended to apply to the mountain vs molehill I fully agree that propulsion is an even bigger mountain
it sounds to me you need enough fuel to slow down and have the main craft orbit Proxima?
Quote from: hop on 08/31/2016 02:33 amQuote from: Robotbeat on 08/31/2016 02:13 amQuote from: hop on 08/31/2016 02:08 amNo, it was preprogrammed, not autonomous....Sure, that works.Except it doesn't work for an interstellar mission. That's the whole point. You cannot apply anything like the approach used by New Horizons or any other real space mission to date.And yes, the disagree was only intended to apply to the mountain vs molehill I fully agree that propulsion is an even bigger mountain But the molehill thing IS applying to propulsion as well.This thing you keep talking about is /trivial/ compared to propulsion. I mean absolutely trivial. It's just ridiculously hard to get there. MUCH harder than we're used to thinking about.
Quote from: Paul451 on 08/30/2016 04:22 amSo if you want to track a moving target (such as an orbiting planet), then - to skew the view - you need to physically move the telescope laterally to the direction of travel.Depends on the field of view of the telescope.
So if you want to track a moving target (such as an orbiting planet), then - to skew the view - you need to physically move the telescope laterally to the direction of travel.
Yes, Proxima b's orbital diameter is only about 7 million kilometers. At about 40,000 million kilometer distance.
Quote from: Robotbeat on 08/30/2016 01:08 pmQuote from: Paul451 on 08/30/2016 04:22 amSo if you want to track a moving target (such as an orbiting planet), then - to skew the view - you need to physically move the telescope laterally to the direction of travel.Depends on the field of view of the telescope.How? It'll already need a wide FoV to observe the entire Einstein ring around our sun. To change what object is "projected" onto that Einstein ring (or to track a moving object), it will need to physically move (not tilt, but move) laterally....
Image the entire system as an object. The distance b orbits from Proxima is about 1000 Earth radii.Kepler, for instance, has 42 CCDs each ~2000x1000 pixels. So if the whole orbit was in view, you should have many pixels of resolution of the exoplanet. And of course, we could have 1000 CCD arrays instead of just 42.
Quote from: Robotbeat on 08/31/2016 03:30 pmImage the entire system as an object. The distance b orbits from Proxima is about 1000 Earth radii.Kepler, for instance, has 42 CCDs each ~2000x1000 pixels. So if the whole orbit was in view, you should have many pixels of resolution of the exoplanet. And of course, we could have 1000 CCD arrays instead of just 42.That's not how Einstein rings work.At 700AU (so the focal angle is far enough away from the sun for the Einstein ring to be visible beyond the sun's atmosphere) the FoV of an Einstein ring is around 1.5 arcseconds. AIUI, in practice, the useful FoV will be about a tenth of that: 0.15 arcseconds.A 14 million km orbit in the target system will have an apparent width of around 13 arcseconds. Nearly two orders of magnitude too wide to image from a single location.