while generally agreed,the Armstrong / Scott Gemini # flight comes to mind where a degree of manual intervention and decision making is required. other situations, e.g Appolo 13, also might qualify as "outliers". Just a thought.
Quote from: ELinder on 11/09/2018 01:58 pmAll this faith in the systems may be justified, but I still think it's a case of "hope for the best but plan for the worst". There's a reason so much training is devoted to emergency procedures. If it were as foolproof as ideally hoped, why not just remove all the displays and controls completely from the manned spacecraft and use that weight savings to bring more cargo mass into orbit?What exactly do you propose the astronauts would be physically (or fly-by-wire) able to do in an emergency? It's not like there's a yoke and stick tucked in the armrest they can pop out and "wing it". The calculations required to burn the thrusters so the situation does not become worse are quite frankly "impossible" for a human to do on the fly (i.e. in an emergency). Other than that, I just can't come up with any other instances of "human capable" tasks that exist on this capsule that fit this emergency procedure list. I would love to hear your insight as to what they could be though, I am certainly not an astronaut who has trained in this capsule, or even anything resembling an expert, or lay-person at best!
All this faith in the systems may be justified, but I still think it's a case of "hope for the best but plan for the worst". There's a reason so much training is devoted to emergency procedures. If it were as foolproof as ideally hoped, why not just remove all the displays and controls completely from the manned spacecraft and use that weight savings to bring more cargo mass into orbit?
So many younger/newer folks are simply in denial.Humans will *not* fly these spacecraft. They are simply passengers.Well educated passengers, but JUST passengers nonetheless.
Yes and no, humans will fly these spacecraft but it wont be the astronauts it will be the programmers that make the code that governs the computers on this spacecraft.
Quote from: clongton on 11/10/2018 01:57 pmSo many younger/newer folks are simply in denial.Humans will *not* fly these spacecraft. They are simply passengers.Well educated passengers, but JUST passengers nonetheless.Not saying this defensively as I 100% agree with this, but as a younger space travel amazing people, fresh and spry at 26, I don't see the argument that the passengers MUST be in control. ...
So my whole task was to basically try to figure out what the rotation was, null it, establish an orientation, and then spin. But the problem with the station is that it has unequal moments of inertia.I remembered from my days in Cambridge that there were these things called Euler's equations, which you use to describe a rigid body that has unequal moments of inertia. So if you think of the Mir having an a-axis, a b-axis, and a c-axis, there are unequal moments of inertia about those. When you put a spin on the station, generally the spinners are not going to stay about one of these axes, unless it's the eigenvector of the inertia tensor.The problem is sufficiently complicated, at least for my intellect, and I tried to solve this problem using Mathematica in orbit.I started off basically with these matrices. I said, "What are the rotation matrices about the three axes?" I use those because that's the convention that's used by Euler. I must say -- this is no ping on Mathematica -- but Mathematica does not implement Euler angles the way I was taught them at Cambridge. Taking those matrices allows you to transfer from the rotating body back to space. In Euler's equations -- if you look at the definition of an inertia tensor -- the angular momentum J is just the sum of the moment times the momentum of each particle.The point here is that it's a tensor, and it combines with the angular momentum vector in the rotating body frame to make the total angular momentum. And only when it's the eigenaxis -- so you have lambda here -- do you actually have a rotation that doesn't change to another axis. That, for me, as an astronaut on the Mir, meant that if I could only find the eigenaxis of the inertia tensor, and get us spinning about it, we'd stay stable and the arrays would still point. That was my goal. But there is another trick in this, and that is, if you have an irregular body with three unequal moments of inertia, the middle axis of inertia is unstable. And it's not unstable in a precise sense, but there's a momentum exchange between the other axes. You have to be right on it -- infinitesimally close -- for it to remain there.The Euler equations are solved (with) the highlighted equation in the body frame, and then using those Euler matrices to transform it back to the space frame. I knew these equations were what I needed to solve this problem. But every time I set this up in Mathematica, we would lose power again. I remember a time when I was setting these equations up, and the whole station powered down, the laptop went dead, and I lost what I had done; and we were once again, for thirty hours, charging around trying to get the station back into shape again. I ended up resorting to what I had in space, which was zero gravity, and making a model and spinning it.
It turned out that the c-axis was the middle moment of inertia, which is unstable; and so we did this flip-flop that actually reduced the power we had and made it much more difficult for us. If -- and this was not the case -- we'd been able to rotate in roll, about the a-axis towards the sun, and then those solar arrays -- as long as they weren't totally shadowed by each other -- would have been able to generate power stably in that configuration. And the a-axis is actually the smallest one of them. The a-axis I called about 1; the c-axis is about 1.1, 1.2 in the package; and the b-axis is about 1.3. They're pretty close to each other.These moments of inertia are not well known by the ground control, because the payload on board is being moved around. When Progresses come up and trash leaves, things are being moved around, and they lose track of exactly where all the stuff is on the station. With water moving around on board the station as well -- the condensation on the walls, maybe seven tons of water moving about on the walls -- it was very hard to know exactly what these moments were. But I did figure out, from a number of attempts to do this, that the c-axis is, indeed, the most unstable axis to use -- and each rotation is roughly twenty minutes. But the arrays lost track of the sun, and we ended up upside-down in relation to the sun. And then the cycle would continue. And this whole flip to upright again, where the solar arrays are once again able to track perpendicular to the sun, became the way we set it up. This cycle would last about eight or nine hours.You can see that this is not the ideal way to spin a station. There was just enough energy -- in spite of this transition that was going on, where it flipped upside down -- there was enough energy obtained by those arrays to let us do the repairs over a thirty-hour period, reestablish the computers on the base block, turn the engines on, stop the station dead, and then slowly rebuild the attitude control and navigation platform.
A safe system must be able to deal with comms and telemetry outages, so there has to be the possibility of autonomous control. And what about when SpaceX flies to Mars?
Apollo 12 after the lightning strike may be a better example....yes S5 guidance kept them flying but Conrad controlled the decision to not abort. Will such decision making still be retained by the commander?
On Crew Dragon just about the only thing the crew can do to intervene in the computer-controlled ascent is to use the manual abort capability. Which will end the flight.
The avionics can be wrong.(Whether or not the crew can do anything about it is another question entirely)New reports out of the Lion Air crash suggest that the failure may have been a new safety system gone wrong due to bad input from a sensor. The crew apparently did not know about this system nor about how to turn it off if needed). (google Lion Air Crash for numerous references).
