9 ms·

To be fair, the capability for autonomous computing back in those days was far more constrained, especially given launch lift costs and the mass of computing, s

To be fair, the capability for autonomous computing back in those days was far more constrained, especially given launch lift costs and the mass of computing, so you had to have humans

skissane 6y ago
What you say may well be true for Apollo. But for the Space Shuttle, even from the first mission back in 1981, most of the flying was done by the computers. There were only a few steps which had to be done by humans – one of which was flicking the switch to lower the landing gear. NASA could have fully automated it, but they made a decision not to. Part of the reason was a culture which wanted to keep a human in the loop rather than having everything under computer control. It was only after the Columbia disaster that they actually enhanced the Shuttle to enable it to be fully autonomous and perform landing and re-entry without a crew. If the Shuttle was damaged and re-entry was too risky, then the crew would shelter in the ISS and the Shuttle would attempt to re-enter and land without the crew onboard – best case scenario, the Shuttle lands intact; worse case, it breaks up on re-entry while the crew remain safely behind on the ISS. They never had to use this autonomous re-entry/landing capability, however.
- dnautics 6y ago
  That's fair, after all the contemporaneous buran was quite capable of running completely autonomously. (N==1)
  - m4rtink 6y ago
    And not only that, you could mount a couple turbojet engines on it and fly it like an airplane! :) https://www.buran-energia.com/bourane-buran/bourane-consti-reacteur.php https://www.buran-energia.com/bourane-buran/bourane-consti-r...
- vibrolax 6y ago
  Sorry for not providing a link to the document giving the procedure for automated shuttle contingency operation. IIRC, there was no plan to de-orbit and land the orbiter. The plan was to ease it away from ISS, open the cargo bay doors, execute the de-orbit burn, and orient the vehicle to break up and burn as completely as possible. Of course, the digital autopilot was capable of performing an automated landing. It was always done manually, ostensibly to maintain pilot proficiency and minimize the reaction time should the autopilot misbehave.
  - skissane 6y ago
    > Sorry for not providing a link to the document giving the procedure for automated shuttle contingency operation. IIRC, there was no plan to de-orbit and land the orbiter. The plan was to ease it away from ISS, open the cargo bay doors, execute the de-orbit burn, and orient the vehicle to break up and burn as completely as possible. That's not true. They implemented "Remote Control Orbiter" (RCO) capability which was based on a special "In-Flight Maintenance (IFM) cable". See this NASA presentation for some details: https://ntrs.nasa.gov/api/citations/20070019347/downloads/20070019347.pdf https://ntrs.nasa.gov/api/citations/20070019347/downloads/20... The plan to intentionally trigger the orbiter to break-up on re-entry was really only for STS-114 – that was the first post-Columbia mission, but RCO and IFM cable weren't available yet. From the next mission, STS-121 onwards, the IFM cable was stored on-board the ISS, so the plan was to use RCO to re-enter and land under remote control. (I'm not sure what the plan was for STS-125, which I believe is the only post-Columbia non-ISS mission – they couldn't use the IFM on the ISS since it wasn't reachable from their orbit; I don't know if NASA had a second IFM cable to use for it.)

To be fair, the capability for autonomous computing back in those days was far more constrained, especially given launch lift costs and the mass of computing, so you had to have humans

skissane 6y ago

What you say may well be true for Apollo. But for the Space Shuttle, even from the first mission back in 1981, most of the flying was done by the computers. There were only a few steps which had to be done by humans – one of which was flicking the switch to lower the landing gear. NASA could have fully automated it, but they made a decision not to. Part of the reason was a culture which wanted to keep a human in the loop rather than having everything under computer control. It was only after the Columbia disaster that they actually enhanced the Shuttle to enable it to be fully autonomous and perform landing and re-entry without a crew. If the Shuttle was damaged and re-entry was too risky, then the crew would shelter in the ISS and the Shuttle would attempt to re-enter and land without the crew onboard – best case scenario, the Shuttle lands intact; worse case, it breaks up on re-entry while the crew remain safely behind on the ISS. They never had to use this autonomous re-entry/landing capability, however.

dnautics 6y ago
That's fair, after all the contemporaneous buran was quite capable of running completely autonomously. (N==1)
- m4rtink 6y ago
  And not only that, you could mount a couple turbojet engines on it and fly it like an airplane! :) https://www.buran-energia.com/bourane-buran/bourane-consti-reacteur.php https://www.buran-energia.com/bourane-buran/bourane-consti-r...
vibrolax 6y ago
Sorry for not providing a link to the document giving the procedure for automated shuttle contingency operation. IIRC, there was no plan to de-orbit and land the orbiter. The plan was to ease it away from ISS, open the cargo bay doors, execute the de-orbit burn, and orient the vehicle to break up and burn as completely as possible. Of course, the digital autopilot was capable of performing an automated landing. It was always done manually, ostensibly to maintain pilot proficiency and minimize the reaction time should the autopilot misbehave.
- skissane 6y ago
  > Sorry for not providing a link to the document giving the procedure for automated shuttle contingency operation. IIRC, there was no plan to de-orbit and land the orbiter. The plan was to ease it away from ISS, open the cargo bay doors, execute the de-orbit burn, and orient the vehicle to break up and burn as completely as possible. That's not true. They implemented "Remote Control Orbiter" (RCO) capability which was based on a special "In-Flight Maintenance (IFM) cable". See this NASA presentation for some details: https://ntrs.nasa.gov/api/citations/20070019347/downloads/20070019347.pdf https://ntrs.nasa.gov/api/citations/20070019347/downloads/20... The plan to intentionally trigger the orbiter to break-up on re-entry was really only for STS-114 – that was the first post-Columbia mission, but RCO and IFM cable weren't available yet. From the next mission, STS-121 onwards, the IFM cable was stored on-board the ISS, so the plan was to use RCO to re-enter and land under remote control. (I'm not sure what the plan was for STS-125, which I believe is the only post-Columbia non-ISS mission – they couldn't use the IFM on the ISS since it wasn't reachable from their orbit; I don't know if NASA had a second IFM cable to use for it.)