NEO Mining Question

I see two classes of feasible scenario for mining NEA. I would
appreciate people's feedback on which they think is most plausible
and any flaws in logic, physics or engineering that I've made. Also,
I don't consider it feasible to move an entire asteroid. The kinetic
energy required for a mass of that huge size is mind moggling.

The first scenario is to send out a mining ship, process the
material and either send the ship and material back, or keep the
mining facility there doing its mining and refining, while the ship
heads back to earth-space.

The second scenario is one that I'm leaning in favor of. Send out a
ship with a mining facility, seeds of a mass driver and the seeds of
a power facility. Establish the facility, extract the material
needed to construct the main mass of the mass driver and the solar
power facility, integrate the earth-manufactured parts for these
devices, and then commence shipments to earth-space. After a while,
the crew departs back for earth.

Obviously, this second proposal is more complex. I envision a small
solar furnace, extracting the bulk materials required for the
manufacture of the systems. This shouldn't be too complex. It
wouldn't need to capture gases, so minimal storage would be
required. It has only one mission: process enough steel, etc to get
the system functional. The rest of the mined material will be
refined in earth-space.

So, here are my technical concerns. Can a mining/excavating system
run autonomously? I'm envisioning a territory is staked out and the
excavating units are bounded within that territory. They excavate
under some AI heuristic, head back to the processing center, release
their load into the mass driver processing unit, then head back to
mine more material. The dropped off material is weighed to a
standard mass, an orbit is computed, a variable velocity of launch
is applied.

So, how complex a system do you think we can leave unattended with
present day techology? I guess this question applies to self-repair
capabilities. I don't know the state of the art in this engineering
field. Can anyone help? No future guesswork please. I can imagine
things too, but for now, I don't really know what is capable.

Another problem is if the the mass driver is attached to the NEO
then as the NEO progresses in its orbit the mass driver will become
oriented in the wrong direction to fire a payload to earth-space. A
solution I foresee is that downrange of the mass driver is a
particle beam cannon which applies the correct amount of energy to
the "in-flight" payload to alter its course. This seems problematic,
but I'm not sure. Or the mass driver can be pivotable. So it is off
of the NEO and the payload is sent to it on conveyor belt. The mass
driver is oriented to allow the payload to achieve proper orbit. But
how do we mitigate the forces of the payload launch which are acting
against the point of attachment between the mass driver and the
NEO? Perhaps we don't attach the mass driver to the NEO. The
payloads are lobbed up to the mass driver but we'd still be left
with the mass driver changing its postion relative to the NEO. So it
would need some propellant to bring it back into position. I tend to
favor the mass driver being off of the NEO, because now the launches
aren't confined to the plane of the NEO. But I'm still left with a
lot of feasability questions. I'd very much appreciate people's
thoughts on the specifics of making such a scheme work.

Now when the payloads are appraoching earth-space, the question
becomes how much has the course varied over the weeks, months, or
years of transit? I've seen reference to a lunar mass driver being
able to hit a target within 10+ meters. But over the course of tens
of millions of kilometers, the error may grow to a 100+ kilometers,
so, now what? If the particle beam guidance system works, then how
about a series of accelerators stationed around the perimeter of the
100 km zone of error. The propellant needed to keep these
accelerators on station should be relatively easy to provide
considering their proximity to earth-space. Radar guidance could
establish the trajectory of the incoming payload, and the correcting
force vectors can be applied from any number of the accelerators. At
the same time they could be applying energy to slow the payload's
velocity. The payload is thus guided to a mass catcher and taken to
the Materials Processing Facility. Is this achievable? Or is this
science fiction at the present? I know we don't have this equipment
now, but is the technology there yet?

I'm really drawn to the efficiency of mass driver propulsion
compared to the inefficiency of rocket propulsion but I see it as
really pushing our technological limits. How much of what I've
proposed is within out means, and how much am I way off base on?

One last thought, I'd rather send the NEO material to earth
incrementally, rather than moving the whole mass of the asteroid,
consuming its mass for propulsion and not seeing the material reach
earth for many years. At least this way material is reaching earth
incrementally, even though it creates all sorts of guidance problems.

Thanks for your feedback.