Starship Asterisk*

I think computer automation is largely taken for granted. Everyone has had a bad experience with a computer bug so it's easy to get caught in the idea that computer bugs are normal rather than the exception. I mean sure, there are poorly tested software out there but if you have a dedicated group of engineers working for years on something, they can make it amazingly sophisticated. A highly sophisticated computer system is not something than many people are able to appreciate because it happens silently. Indeed, if the user is able to forget the computer is there, it's probably a sign of great engineering.

JohnD wrote:Yes, I would agree that even in "Apollo 13" the navigation calculations and recommendations appear to have been done on the ground, while the piloting was done by the capsule crew. Is that what you meant?

But it's said that the computer on board the Apollo lunar lander had the capacity of a small handheld calculator, and clearly the crew of that had great (complete?) piloting and navigation autonomy. Computing power has vastly increased while size has decreased, so would at least as much autonomy apply to a Mars lander, let alone a the new Orion?

I'd expect most future spacecraft to have no control stick. The Apollo 13 capsule was still flown manually (which even with a very skilled pilot, was still nearly impossible, and nearly a failure... certainly there was a good deal of luck involved in the success of that mission). Were an identical accident to occur with a modern control system, nobody would have their hand on a stick, nobody would be directly controlling thrusters. Course adjustments and attitude control would be entirely automatic.

Hard to say how a lander would be designed. Probably, full automation would be the norm. An aerodynamic lander might offer more manual control than a rocket lander. Neither would probably offer any manual control during the earlier part of the re-entry process, but it's possible that some manual input would be accepted in the final landing stages. In the long run, though, not even that.

Chris Peterson wrote: Note, also, the important difference between "autonomy" in the sense of making command and navigation decisions, and "autonomy" in the sense of actually piloting a spacecraft.

Yes, I would agree that even in "Apollo 13" the navigation calculations and recommendations appear to have been done on the ground, while the piloting was done by the capsule crew. Is that what you meant?

But it's said that the computer on board the Apollo lunar lander had the capacity of a small handheld calculator, and clearly the crew of that had great (complete?) piloting and navigation autonomy. Computing power has vastly increased while size has decreased, so would at least as much autonomy apply to a Mars lander, let alone a the new Orion?

John

Isn't Orion just a capsule ? They need a spider also to land and take off again, right ?

Chris Peterson wrote:

What will happen when it's "Orion, do you have a problem?"

Most likely, everyone will die. Because that's the usual result of a serious problem in space. However, it its a survivable problem, automation will probably help. There's no reason to think that an Apollo 13 type accident today wouldn't be better handled with automation than any human pilot could manage.

Perhaps more likely would be: "Ah, Mars, she is beautiful. Here we come." [Final transmission.] Then we are left on Earth to deal with all of the conspiracy theories and never really find out what happened.

JohnD wrote:Chris,
Not just the, not just soam-in-the-can Gemini astronauts, but the Apoloo crews too, had autonomy. Or was Apollo 13, the film, wrong? Memory tells me that it was one of that crew who adjusted the positioning of the capsule and fired the rockets to re-align them for Earth entry. Or was all that done by Houston Control?

Where did I suggest otherwise?

What will happen when it's "Orion, do you have a problem?"

Most likely, everyone will die. Because that's the usual result of a serious problem in space. However, it its a survivable problem, automation will probably help. There's no reason to think that an Apollo 13 type accident today wouldn't be better handled with automation than any human pilot could manage.

Or, Orion's near Mars orbit and 20 minutes away. Mars Curiosity has autonomy, and did for the landing. Yes, it's a robot, but there could be three humans in there, dying (?) to take control.

No human could perform that landing. If the landing automation failed, there would be nothing humans inside could do about it.

Note, also, the important difference between "autonomy" in the sense of making command and navigation decisions, and "autonomy" in the sense of actually piloting a spacecraft.

Chris,
Not just the, not just soam-in-the-can Gemini astronauts, but the Apoloo crews too, had autonomy. Or was Apollo 13, the film, wrong? Memory tells me that it was one of that crew who adjusted the positioning of the capsule and fired the rockets to re-align them for Earth entry. Or was all that done by Houston Control?

What will happen when it's "Orion, do you have a problem?"
Or, Orion's near Mars orbit and 20 minutes away. Mars Curiosity has autonomy, and did for the landing. Yes, it's a robot, but there could be three humans in there, dying (?) to take control.

John

BMAONE23 wrote:Saw inside the module and would say Moon, OK, beyond to Mars or beyond, too claustrophobic.

No one is going to be stuck in the Orion capsule during a flight to Mars. From the Orion Wikipedia page...

While NASA offers that the Orion capsule will be a part of some future manned Mars mission, with only 80 cu. ft. of living space per crew member, it is clearly only intended as a "near Earth delivery vehicle", to deliver astronauts in the vicinity of Earth to some yet to be designed interplanetary transfer ship, designed to transport future Mars astronauts on their 16 month round trip to Mars.

Rob

ta152h0 wrote:subject of a giant discussion in the office that changed the software on the 777 flight control laws that put the pilot back in the loop as " final authority "

These sort of discussions will continue for a long time yet.

It's worth noting, however, that the pilot of a 777 has no direct control of the flight surfaces. He may be the "final authority" in certain senses, but it's still the electronics that are actually flying the plane. And they will not allow certain things to happen, even if the pilot tries to do them, and that cannot be overridden.

Saw inside the module and would say Moon, OK, beyond to Mars or beyond, too claustrophobic.

subject of a giant discussion in the office that changed the software on the 777 flight control laws that put the pilot back in the loop as " final authority "

https://www.youtube.com/watch?v=Y1FKAIrb0fQ

I thought I would mention that there is a typographical error in the explanation in which it currently states "United Launch Aliance Delta IV Heavy rocket" but the 'Aliance' should be 'Alliance'. I first thought 'Aliance' may be the US spelling but when I looked at the links in the explanation it was clear that it should be 'Alliance'. I appreciate that it may be obvious it is a typographical error but it could cause confusion in future when the APOD is still in the archive but the links are no longer live.

PS. The explanation currently states "Friday at 7:05am ET" but it should perhaps be better to have stated EST not ET, as ET could be EST or EDT without knowing otherwise.

PPS. Orion is my favourite constellation, so I like the name.

Chris Peterson wrote:

Guest wrote:Do people always make the correct decisions? Of course not, but I would rather my life be in the hands of someone with as much to lose as I have, rather than place my life in the hands of the programmer sitting in his recliner watching the latest sitcom while the kids raid the fridge and the dog pisses on the carpet. Just saying is all...

You don't have a choice. The systems have simply become too complex for human pilots. There is no way that a human can react quickly enough and manage all the systems.

People will command the aircraft and spacecraft, but they won't pilot them in any conventional sense. Nobody is going to have their hands on a stick, because there will be no stick. Even automobiles will be this way in the not too distant future. Computers are better at it, and our systems will be safer for it.

And personally, I'd rather my airplane was flown by the software and hardware created by a large team of expert engineers than by a single pilot who might be having a bad day.

But hey, even if you aren't a stoned programmer, you'll probably still enjoy this scene from the (very funny) short film, Thumb Wars: The Phantom Cuticle ...
https://www.youtube.com/watch?v=XlXXoq7N4UQ

http://en.wikipedia.org/wiki/Thumb_Wars

Guest wrote:Do people always make the correct decisions? Of course not, but I would rather my life be in the hands of someone with as much to lose as I have, rather than place my life in the hands of the programmer sitting in his recliner watching the latest sitcom while the kids raid the fridge and the dog pisses on the carpet. Just saying is all...

You don't have a choice. The systems have simply become too complex for human pilots. There is no way that a human can react quickly enough and manage all the systems.

People will command the aircraft and spacecraft, but they won't pilot them in any conventional sense. Nobody is going to have their hands on a stick, because there will be no stick. Even automobiles will be this way in the not too distant future. Computers are better at it, and our systems will be safer for it.

And personally, I'd rather my airplane was flown by the software and hardware created by a large team of expert engineers than by a single pilot who might be having a bad day.

Guest wrote:Just saying is all...

All you've really done is commit a straw man fallacy.

Chris Peterson wrote:

BDanielMayfield wrote:

Chris Peterson wrote:And that's why there are still pilots on airplanes, although in reality, the number of accidents that would be avoided be eliminating cockpit errors probably exceeds the number avoided by a pilot taking effective emergency action. Computers are better pilots than humans.

Unfortunately, when the automated systems replace a highly skilled, highly trained, and experienced pilot... you are still left with a human in the cockpit. That is the programmer (or team) who actually wrote the code that controls the systems. None of whom have their 'butt in a sling' when something goes wrong. Maybe they were hung over when they wrote the key subroutine, failed to initialize a variable because they had to sneak off for a third cup of coffee before 8AM, or had one hit too many on that joint at lunch time and decided to snooze the afternoon away instead of testing all of the possible failure conditions. No. If it were me, I want that person on the controls knowing that we all live or die together. Do people always make the correct decisions? Of course not, but I would rather my life be in the hands of someone with as much to lose as I have, rather than place my life in the hands of the programmer sitting in his recliner watching the latest sitcom while the kids raid the fridge and the dog pisses on the carpet. Just saying is all...

BDanielMayfield wrote:

Chris Peterson wrote:As with most aircraft, pilots are a liability.

The passengers of the fight that had to make an emergency landing in the Hudson River sure didn't feel that way about Captain Scully and his co-pilot. Another of the many examples of pilots with the "right stuff" was the crew of Apollo 13.

And that's why there are still pilots on airplanes, although in reality, the number of accidents that would be avoided be eliminating cockpit errors probably exceeds the number avoided by a pilot taking effective emergency action. Computers are better pilots than humans.

Automated flight is fine when everything is clean and green, but unforeseen accidents happen, and when they do it helps to have smart, highly skilled, adaptable people on hand.

Actually, automated systems are generally much better than humans when things go wrong. Human pilots are fine when everything is clean and green.

Anyway, most modern high performance aircraft (and this presumably includes future spacecraft) physically cannot be flown by humans. They are not aerodynamically stable, and must be flown by computer. All the pilots do is basically point them in some desired direction. In the event of an automation failure, they become rocks, and there's nothing the pilot can do about it.

BDanielMayfield wrote:

Chris Peterson wrote:As with most aircraft, pilots are a liability.

The passengers of the fight that had to make an emergency landing in the Hudson River sure didn't feel that way about Captain Scully and his co-pilot.

That was an exceptional and rare circumstance. It doesn't make the argument that sophisticated computers can fly better than human pilots untrue. Not being familiar with the operation of a large aircraft such as the one Sully ditched in the Hudson myself, I wonder how much assistance the computer offered during the effort. It is not as if the computer failed or turned off completely once the bird strike occurred.

Chris Peterson wrote:As with most aircraft, pilots are a liability.

The passengers of the fight that had to make an emergency landing in the Hudson River sure didn't feel that way about Captain Scully and his co-pilot. Another of the many examples of pilots with the "right stuff" was the crew of Apollo 13.

Automated flight is fine when everything is clean and green, but unforeseen accidents happen, and when they do it helps to have smart, highly skilled, adaptable people on hand.

Bruce

I have a question.
Is this the rocket system that will take marsonauts to Mars and bring them back to Earth?
How do engineers expect marsonauts to survive the five month journey to Mars and land them safely on that planet?
Because of the immense material and biological demands of a journey to Mars only a one way (non-return) trip to the red planet may be feasible. The mission may fail to include a return trip to the Earth. This does not mean that the marsonauts would be stuck on Mars..they can always catch a return trip back to Earth with the help of extraterrestrial aliens
Relying on rocket power alone to land human explorers on Mars and return them to Earth may not be possible due to the overwhelming material considerations of rocket propulsion - the issue of carrying fuel for the return trip. An alternative system must be devised.
Why not fly them to Mars
The space vehicle would be designed with wings so that it can be launched from a flying carrier at an altitude of 50 km. The Mars lander would simply fly into space and out into a Mars trajectory using both the lift of wings and rocket propulsion. On arrival to destination the Mars lander would employ wing dynamics to plunge through the Martian atmosphere and navigate for a proper landing site - preferably in the polar region of Mars. The Mars lander would have to be designed with skis so that it lands on ice in the polar region of Mars. After the mission is completed the Mars lander uses rocket aided propulsion to lift above the Martian atmosphere and return to Earth for a landing somewhere in the Arctic circle - the skis.

Guest wrote:

hoohaw wrote:If this is possible, it may be possible, some day, for people to actually go to the Moon! Hard to believe!

Who was Neil Armstrong???

He was on first.

Guest wrote:Just a quick FYI...

A nautical mile is a unit of measurement used by navigators in shipping and aviation. It is the average length of one minute of one degree along a great circle of the Earth at the surface. One nautical mile corresponds to one minute of latitude. Thus, degrees of latitude are approximately 60 nautical miles apart. By contrast, the distance of nautical miles between degrees of longitude is not constant because lines of longitude become closer together as they converge at the poles. In addition to being used in navigation and aviation, nautical miles are also used polar exploration and international laws and treaties regarding territorial water limits. Also, for reference, there is no metric conversion the for 360 degrees of a circle/sphere. And before someone mentions it, PI would still remain 3.14159265...... because it is a ratio and independent of the unit of measure (the units cancel)

The original reason the metre (or meter) was made the length it is, was so that the shortest distance along the Earth's surface, from equator to pole, could be expressed as 10,000,000 m, or 10,000 km. These days, the definition of the metre has been refined, and the nautical mile has been redefined in terms of the metre, so that neither are defined precisely by the Earth, but the circumference of the Earth is still pretty close to 40,000 km, or 21,600 nautical miles, no matter how you measure it.

Guest wrote:

Chris Peterson wrote:The standard metric (SI) angle unit is the radian.

Radian describes the plane angle subtended by a circular arc as the length of the arc divided by the radius of the arc. A Radian can be described in the absence of a 360 degree circle whereas PI by definition requires the full circle (circumference)...

Your point escapes me. You seemed to be suggesting there was no metric conversion for degrees. That isn't the case.

Chris Peterson wrote:The standard metric (SI) angle unit is the radian.

Radian describes the plane angle subtended by a circular arc as the length of the arc divided by the radius of the arc. A Radian can be described in the absence of a 360 degree circle whereas PI by definition requires the full circle (circumference)...

ta152h0 wrote:I remember Gus Grissom expressing an opinion on this " automatic " stuff and it wasn't pretty.

Well, he came from the world of old school test pilots. His view wasn't surprising.

But today, our automatic systems are vastly superior to humans the great majority of the time. That is reflected in the degree to which automation is taking over both commercial and military aircraft, as well as spacecraft. Even where people fly them, they're really just providing a bit of direction for what are essentially automatic control systems.