TECHNICAL ISSUE: COMMUNICATIONS TIME LAG
One reason the idea of teleoperated robots has been limited to use on the moon (see LunaCorp's Robots in Cyberspace) is that the lag between sending a command to a robot and seeing the result could easily be 16 minutes if the comet core is in Earth orbit but on the other side of the sun. On the other hand, if a comet core is in a near-Earth orbit, it is likely to be going the same direction as the Earth, and so will be in the neighborhood of Earth for a long time. If the robots can land, extract water, and take off after only a few months, they might be considerably closer to the Earth during the rendezvous. In that case the lag might be only a couple of minutes.
More to the point, the mission for moon robots is open ended - to wander around taking samples. This requires a large number of actions on the part of the robot operator. An ice miner, or the other hand, has a more limited task. It lands, pushes away rubble, slowly lowers a hot plate into the ice, and collects the low-pressure steam coming from the hole into a balloon.
On the moon there is a danger that falling could damage a robot - especially a large robot. There is no such danger on a comet core. Hence on the moon, but not on a comet core, it is important to move carefully. An asteroidal robot could move by jumping, and if it didn't like where it landed, it could jump again. Communication for locomotion will be simpler on a comet core.
Another way to look at the communication time lag is as a slowing down. An action that would normally take two seconds now takes sixty times as long. A task that a person could do in one day would take a robot two months. Yet if the task is simple enough, that may be adequate. The alternative is not one day versus two months, but two months versus the ten years (and high cost) it might take to develop special-purpose semi-autonomous robots.
Still, operating a robot at such a distance is likely to be extremely frustrating. It is important to minimize the number of operations which need to be performed at the comet core and do as much as possible later on in Earth orbit. If work at the comet core can be easily automated (not require much development time), then it would be a good thing to do so. With regard to space development as entertainment, the difficulties of the teleoperator do not need to be communicated to the audience to be entertained; all the time lags can be edited out.
To read a defense of autonomy, click here.
If you see any large holes in the above arguments, we would appreciate hearing about them.
