Lighter Kliper could make towed trip to ISS

[realtime]

So Parom is a fully automated system?  What's to stop them from slamming into the ISS proper, like the Progress did with Mir?  Maybe I've seen far too much horrible computer code.  Fully automated systems scare the hell out of me.  Then again, no system is without risk.

There should be different set of standards for expendable cargo and humans.  This should apply at the architectural level.  Loss of cargo, while undesirable, can be accepted.  So, when designing cargo systems a certain amount of redundancy can be dispensed with in the name of economics.  When designing systems that will carry humans, no such corners can be cut.  By using Parom for both cargo and humans, the Russian designers have placed themselves in a position where they must "waste" design margin on cargo or underdesign failsafe systems for humans.  

This was a hard lesson that NASA learned from STS, but I believe it's finally sunk in.  It may indicate a certain level of desperation (or avarice) on the part of the Russian space bureau that they would willfully ignore this lesson.  Just shows they're human too.

It's inevitable that compromise will adversely affect the design.  I hope it doesn't produce any fatal flaws.

[lmike]

I may have jumped the gun with the details, the new drawing still shows some RCS thrusters in the nose of Klipper, perhaps it's still meant to do the prox ops with Parom, but where would the fuel tanks be?  They used to be in the SM module in the back, AFAIK.  Too few details... The Parom could be teleoperated (but even then, the objections stand), I guess, but Russia has very limited communication coverage capability. Only short windows of telemetry/command uplink while it passes over  their territory.  

Another thing with this new scheme, is that the original Klipper promised same day docking with the station as an improvement, or independent flight with tourists for up to 5-7 days.  (the 'nasty' thing about the Soyuz is that it has to do some fancy orbit legwork to conserve fuel and can only dock with the ISS on the 3rd day after the launch, hence the orbital module is there to relax, at least)  Now, this thing launches and sits waiting there...  with people inside... for how long before getting to the station?

(not saying it's impossible, just seems overly complicated)

[publiusr]

I wonder if ATV could be launced atop UR-500 Proton. Too far north? Wasn't a bigger hydrogen upper stage proposed for Proton? Or has Angara usurped everything as Proton's replacement?

[lmike]

publiusr - 23/11/2005  12:42 PM

I wonder if ATV could be launced atop UR-500 Proton. Too far north? Wasn't a bigger hydrogen upper stage proposed for Proton? Or has Angara usurped everything as Proton's replacement?

Some people advocated (and still do) just the development of the cryo LH2 upper stage for the Proton, the smaller cousin of it, built by the very same Khrunichev that has already been built and *flown* on the Indian GSLV, *instead* of sinking billions into Angara.  Some heavy lifting...  In a couple of years instead of a decade... But Khrunichev (the main pusher of the Angara) has the President's ear... politically speaking.  The ATV is, AFAIK, is tied to the Arianne.

[JonClarke]

"So Parom is a fully automated system? What's to stop them from slamming into the ISS proper, like the Progress did with Mir? Maybe I've seen far too much horrible computer code. Fully automated systems scare the hell out of me. Then again, no system is without risk."

There have been over 100 successful dockings with Progress, not one primary docking has failed.  the only failure was a Progress under human control, in that case the operator was not properly trained for the proceedure and it was an unusally unfavourable configuration.

There is no reason whatsoever to think that automated dockings will not be routine and save.  As with Progress one can assume that there will be a manual over ride available.

Jon

[lmike]

Right, the collision with Mir was due to a tired and poorly trained crew using the TORU manual control system on the ISS (for some weird reason).  You are also right that *automated* , not *automatic* (a significant semantic distinction) dockings with the KURS system have been successful in the past with large and stable targets such as space stations (smaller targets such as the Klipper may spin slightly, may drift out of sight of the receiver, and the target beacons cannot be spread out as much)  

Just to expand.  The very same KURS, on the Soyuz, however, has failed in the past, such as the recent TMA-5 when the approach speed chosen by the computer was too great and manual control had to be taken to prevent collision (and it'd only take one to kill the occupants of the Klipper)  However, as far as this topic is concerned, the Progress/Soyuz experience has no bearing on the Parom, or space based tugs in general.  It is not even close to an autonomous vehicle a tug would need to be!  

Their long range approach is done *teleoperatically* except that the operator(s) are on the ground down in Korolyov (TsUP).  They (the ground in TsUP) get telemetry and upload commands for the thrusters (whenever the TsUP has a comm 'window'...).  The automated part is during the prox ops with the station when the KURS system kicks in (at ~5km), within 150m it has to be monitored by the ground.  In addition, the Progress is a one shot system, while a tug is a multi-use long duration vehicle by definition that has to live on orbit.  The KURS system is not a good candidate for the Parom, as it requires an active (transmitting) radio beacon on the target, and the target has to be stable, .... and recalibration for differen targets.  A tug's target's close range beacon system would need to be passive and visual, and robust enough to deal with different sizes, masses, positional changes, and reflective properties... And to top it off, the KURS is manufactured in Ukraine and Russia has been salvaging the units from the ISS via the Soyuz and re-conditioning them, so it's not a good canditate for the future vehicle, perhaps they can switch to the ATV's laser-based system in the future.

To summarize: we have not yet seen even a technology demonstrator needed for any tug (some of the ATV components may come close to being autonomous)  Let alone a robust enough one to do this scheme of launching people to the station.

(personally, with the amount of non-standard target conditions a tug would have to deal with I'd strongly prefer a trained human in the loop)

[publiusr]

Well spoken. I still don't fully trust robotics.

[Avron]

lmike - 11/12/2005  3:49 AM

To summarize: we have not yet seen even a technology demonstrator needed for any tug (some of the ATV components may come close to being autonomous)  Let alone a robust enough one to do this scheme of launching people to the station.

(personally, with the amount of non-standard target conditions a tug would have to deal with I'd strongly prefer a trained human in the loop)

and still not demonstrated. The risk is too high, give me a human anythine, but if you want a real mess, give it to a machine that has just been Zapped by some cosmic ray or two..

[JonClarke]

I would very surprised if the would not be a human in the loop with Parom dockings even if the baseline procedure would be automated.  This would be both unprecedented and unjustified.

Jon