On Earth, before a mineral or petroleum resource is mined/extracted, the deposit is delineated and evaluated.
Briefly, the process involves sending a some geologists and some drill rigs and their operators to a deposit and drilling holes through the deposit on a predetermined grid pattern. The drill cuttings or core (depending on the type of drill used) will be logged and samples taken at uniform intervals for assaying.
The logging of the geology will be used to determine the structure and nature of the deposit and the rock surrounding the deposit.
The assays will define the distribution of grade - the amount of metal per tonne in a given region. This helps narrow down the better parts of the deposit to mine, should it prove profitable.
A mine design is then done and evaluated and if profitable it can then be mined, following approvals.
For the Moon, or elsewhere in the solar system, this is unlikely to happen because of the expense involved - see edit at the end.
Technically, getting any equipment to the Moon is possible. It may require establishing a workshop on the Moon and sending the equipment as partially pre-assembled parts and then assembling them on the Moon, but it's expensive.
Satellite or drone data can give an indication of lateral grade distribution but not the vertical distribution that is required for depth of mining - see comments later.
If helium-3 is mined on the Moon it will essentially be a scavenging exercise.
If Wikipedia is correct about helium-3, the grade of helium-3 is very small: 1.4 to 15 ppb (parts per billion) in sunlit areas and up to 50 ppb in shadow areas. By comparison, on Earth, gold can be mined profitably by open pit methods if the grade (concentration) is 5 g/t, or ppm (parts per million). Helium-3 on the Moon is 1000 times as dilute. To mine such a low grade material profitably would require large scale mining & processing.
Popular Mechanics states:
Digging a patch of lunar surface roughly three-quarters of a square mile to a depth of about 9 ft should yield about 220 pounds of helium-3.
Now 0.75 sq mi is 1,942,491 m2, 9 ft is 2.743 m & 220 lb is 99.79 kg.
In this situation, the volume of regolith needed to be mined would be 5,328,641 m3. Using a density for the regolith as 1.5 t/m3 (something like clean sand), the mass of the dug regolith would be 7,992,961 t (8 Mt) and the grade of helium-3 would be 12.48 ppb (mg/t, milligrams per tonne).
The mining of 8 Mt of regolith will require a large fleet of robust mining equipment, particularly if the material is to mined at a reasonable rate.
Lunar Helium-3 and Fusion Power, published by NASA in 1988, on pp21-22 envisaged the use of draglines or ballistic miners to dig several small pits per year.
Most of the helium-3 occurs in sub 50 $\small\sf\mu$m regolith.
The mined material would be screened to discard the plus 4 mm size fraction. The remaining sub 4 mm material would be electrostatically treated and then heated to 700 $\sf^{\circ}$C to release the volatiles.The gas would then be compressed for storage. Use of bucket wheel excavator digging to 3 m depth at a rate of 1258 t/h; 3942 h/year would produce 33 kg of $\sf{^3}$He per year.
Heating the regolith could be done by a solar thermal furnace, 110 m in diameter & 10 m deep.
The evolved gas is subjected to a cooling/condensation process to liquify the different species. The evolved He is and separate the different species. Cooling takes place during the lunar night to make use of the ambient cold. Hydrogen is removed before cooling. Other gases produced by the process include H20, 02, N2, H2, etc. subsequently cooled to 55 K for preliminary isotopic separation and then through a cryogenerator (to 1.5K) to achieve maximum He-3 concentration (99%).
The terrestrial mining experience base is large and should be tapped. A specific analog to lunar regolith mining would be mining of terrestrial mineral-sand deposits. The grade and distribution of "ore" would need to be known which in turn would require assessments ahead of the miner for planning purposes. The low lunar gravity ($\sf{\frac1 6}$ Earth's) may be a problem for the equipment. The machinery may require ballast to achieve the required mass/inertia for mining. Since transportation costs are high, the ballast may have to be supplied from lunar materials.
The document goes on to state that given then usage of energy in the US, at that time, 600 lunar mining machines would be required (p25).
Two significant questions raised in the NASA publication were:
- What is the concentration of He in the lunar regolith as a
function of depth as well as its areal distribution
2. Why is the He abundance in the regolith strongly correlated with
the TiO2 concentration?
Without drill hole data, grade distribution with depth cannot be known prior to mining.
Another complication is the topography of the bedrock is unknown. It cannot be assumed the bedrock is smooth and flat. Undulating and/or jagged bedrock will be problematic for any mechanical digging device such as a drag line, bucket wheel excavator, or a scaper. It will also reduce the amount of regolith that can be mined.
Additionally, the Moon's day and night cycles will affect mining operations - 13.5 days of sunlight, with surface temperatures up to 127 $\sf^{\circ}$C followed by 13.5 days of darkness, with surface temperatures down to -173 $\sf^{\circ}$C. Heating the regolith will only be able to be done during the sunlight period if a solar furnace is used.
Some more information about mining helium-3 on the Moon.
As mentioned in the comments, the feasibility of nuclear fusion as a controllable energy source is not yet proven and mining helium-3 on the Moon as a feed stock would be very expensive and require a lot of effort and would have to compete with other sources of energy.
Edit 2 April 2020 - NASA cost estimate for cost of launching material into orbit
There's certainly a strong argument to be made for using existing materials on the Moon itself to construct a lunar base. NASA estimates that it costs around $10,000 to transport one pound of material into orbit
That's 4536 USD per kilogram.
Edit 21 April 2020
Astrobotic says it's building a "railroad" to the Moon.
$1.2m per kg for heavier items such as robot rovers, Astrobotic will deliver the payload of your choice to the Moon, using its four-legged Peregrine robot lander.
