Man vs. Probe.

[magnemoe]

22 hours ago, Kryten said:

KSP is a video game.

This, its far simpler, also KSP operate with an far higher budget.
Its no huge cost sending an 50 ton ship to Duna. 
If that had been the situation real life we would had an manned Mars mission long ago. 
Add that probes are somewhat limited in that they can do in ksp, however if you can avoid using kerbals like for the do science readings from various locations I prefer probes as they are lighter and you don't have to return them.

[Servo]

I feel that despite a probe's obvious superior cost/scientific abilities, manned missions are necessary for the space program. While robotic missions accomplish more scientific goals, manned missions generate interest in space, engineering, and science as a whole. The "cool factor" of a Mars mission would inspire a massive increase of interest in scientific fields, even ones not directly connected to the space program. Unmanned missions don't generate the same buzz (anyone excited about the LADEE mission?). I'm not saying either type should be exclusively undergone, just that the benefits of manned missions aren't all scientific or immediately obvious.

[tater]

The only reason for crew to go in that flyby is to shorten the time lag for telepresence. As time goes forward, the ability to have more autonomous robots only increases. No one has yet sent anything remotely as capable of self-driving as a google car, for example. Once we have something like that, then the ground crew is picking a target, and perhaps a guesstimate route, then the vehicle just drives there, and perhaps at a substantially higher speed than cm/min. A robot that could move at a slow walking pace autonomously would cover vastly more ground than anything we have ever sent.

[DBowman]

@tater I agree re the self driving. Current rovers are excruciatingly slow, better self drive would take care of that. Probably the remaining opportunity for  humans doing low lag telepresence is poking around at a new site - quickly evaluating, moving, sampling, etc.

This article seems keen on the concept but says we don't have enough experience to understand the benefits.

This Wired article quotes Steve Squyres (lead scientist of Spirit and Opportunity robo duo) as saying a human geologist could do in a week what the two rovers did in five years - better automation would cut dramatically into that advantage - but how far? NASA is doing tests where they teleoperate robots from ISS (why not just teleoperate from 'the office' with artificial delay inserted? it's seems a waste of $$ to do it from orbit)

This NASA one is interesting look at a future roadmap for surface operations but only talks about human+robot in the context of surface EVA. Not sure if thats because they are focused on 'surface operations'.

[tater]

@DBowman, excellent points. The quote from Squyres is particularly useful as it actually characterizes what someone in the rover community thinks their data acquisition rate is compared to a person on the ground. The first reasonable "data" I've seen here on that subject.

[Ten Key]

37 minutes ago, DBowman said:

NASA is doing tests where they teleoperate robots from ISS (why not just teleoperate from 'the office' with artificial delay inserted? it's seems a waste of $$ to do it from orbit)

Test as you fly. Fly as you test. It's the crazy little things you didn't think of that can end up killing you. 

[DBowman]

4 hours ago, Ten Key said:

Test as you fly. Fly as you test. It's the crazy little things you didn't think of that can end up killing you.

Sure, but it sounds like we are in the early stages of proof of concept for man+machine working. The mission they are testing is an Orion at L2 directing a Lunar robot to unroll some radio telescope components (so Orion will 'hang in the sky'), ISS orbits once every 90 mins. It just sounds like a very expensive way to do a 'low fidelity' test, maybe the actual robot is just 'icing on the cake' and incidental to the real on board procedures exercise - but it smells like a publicity exercise. ISS is so expensive you'd hope there was a high 'it can only be done on ISS' bar that had to be jumped over.

I think you'd get good value for money from running competing teams of uni students through real or simulated environments under various team structure and latency regimes. Should you put 'payload specialists' in a can round Mars or generalist astronauts trained up a bit and consulting to the specialists? how does it effect the science vs cost vs risk trade-off? etc A few guys on the ISS isn't going to explore that trade space very fully.