a) Slaves are useful at least as test subjects
b) Loyalists have a nice advantage to form highly autonomous sub hives, and after some technological solutions may be made much better, reliable and loyal. I really want more loyalists in the future
c) We can get raw materials only by extensive mining, that require more diggers, better equipment and demolition works that will create unpleasant seismic activity, but by raiding we may get some refined materials
d) We will not be equal with bipeds even if we'll use only their technology, because we breed faster, our soldiers are much harder to kill, no morale problems and so on
e) We will not use human technology as it is, we will modify it
f) Human technology will allow us to mimic pirates far longer
We really don't need to test much. The bipeds are important now but are at most a minor consideration in our long-term plans and we will likely be removing them once we have an extensive empire. We shouldn't require biped-specific technologies, although practising how to adapt technologies to a specific opponent may be useful even if it is unnecessary...
A commander is perfectly capable of maintaining an autonomous subhive. The only advantage that the bipeds have over us is that we are somewhat reliant upon a hierarchy with commanders and queens being somewhat necessary and time-consuming to replace. Provided that we maintain a high degree of superiority in our defensive technologies, or produce an excess of organisational drones, the bipeds are comprehensively inferior to us.
To conduct mining efforts in safety we need only conduct our-selves in regions that the bipeds do not occupy. By all accounts this is a sparsely populated community and there should be large regions of the planet subject to only minimal surveillance. The biped communities seem somewhat opportunistic and completely evacuating a facility of its raw materials will almost certainly be noticed and draw attention to the existence of a new processing complex in the region...
We are severely out-numbered on every level. Or current forces are minimal, were we less adaptable we would likely have succumbed due to a lack of biodiversity by now. The bipeds have multiple planets, each likely possessing well equipped interplanetary militaries. Beyond that the bipeds would appear to possess an interstellar militaristic empire and the capacity to contact intergalactic entities. We cannot afford to just abandon any advantage that we may have access to. And we know little of the effects that stressful situations may have upon us. We know that we have lost almost all that there was of our former mind due to some form of trauma, possibly that was purely a result of this hive's dormancy but there is no reason to believe that we are immune to perceptual discrepancies when strained much as the bipeds would be.
With our technological proficiency the abilities of a drone should be almost irrelevant. Automated targeting systems can adjust their target more swiftly than any physical construct could hope to manoeuvre. A suitably potent antisensory weapon should render a target nearly ineffective regardless of its resistances. High-energy weaponry should penetrate any mobile defences. armoured predators may well be effective against poorly equipped forces but against a heavily fortified position we would need to have already colonised a significant proportion of the planet. If we must spread across the planet to engage our enemies, then we may as well make colonisation our focus. If we are seeking to engage is open hostilities with anything comparable to our current production capacity then we will need to advance our technology to levels that will not induce ridicule.
Mimicking biped technology and duplicating it are different, all we require is sufficient information on their sensory apparatus to deceive such systems. Beyond that we could learn more from our own reactor than from any number of remote biped colonies.
Additional hives would be an excellent resource, though they should be primarily for raw material and population production. We should require relatively little processing and conditioning facilities and fixed military resources are not currently something that we can secure against detection while maintaining strategic positions.
You are overestimating human detection ability and their involvement with what happening at remote towns.
Strategic value is dependant upon access to items of value to the bipeds. Their actions are somewhat irregular and in order to maintain military value a facility would require access to the surrounding region, which would expose it to visual detection from the surrounding region. Building military facilities relevant to the bipeds is simply too high a risk to our greatest asset and offer no advantage prior to major operations. We should be expanding away from the bipeds, not towards them.
Aviation has a great degree of visibility and a relatively limited capacity for armouring. Artillery consists primarily of autonomous aircraft or arcing projectiles which require sensory data to aim and have a comparatively long flight-time. A sufficiently swift and comprehensive sensory system should be able to locate hostile sensory devices and disable them using antisensory weaponry and detect arcing projectiles and direct a high-speed vehicle to evade them. The primary remaining military threats would be nonsensory, unpredictable autonomous projectiles, which could likely be prohibited with a similar arrangement using more destructive weaponry and are still dependant upon initial sensory data, low flight-time weaponry which would typically require direct trajectories, and weaponry hostile to large regions, which would typically be chemical agents.
The primary threats that the bipeds possess are information and coordination. Our efforts to disarm them should focus upon disabling their sensory faculties and communication systems. Both of these elements can be rendered non-functional by denying the bipeds knowledge of our existence by preventing contact with them.
a) visibility can be negated by developing stealth technologies
b) After AI Virus will strike enemy radars, their air detection will be greatly reduced
c) Why are you so obsessed with armouring? We can allow ourself to lost a lot, as long as we can produce more. Only key units should get heavy Armour (queens, commanders, other high level drones)
d) Human coordination is their advantage? They are far behind us in that regard, while it may be useful to limit their coordination further, making that main priority of the research is a waste of R&D potential
Stealth technologies can be negated by developing sensory technologies. A wide spectrum motion, contrast, and referential sensor is far easier to construct than a multidirectional visual inhibitor and a directional visual inhibitor is of extremely limited value, and that is just a passive sensor. Include technologies such as detecting the magnetic signature of electrical devices within a magnetic field, active electro-magetic reflection sensors, and simple radiant-heat detection and stealth technologies that do not rely upon a significant quantity of expected mass are almost worthless.
Assuming that the virus is completely successful seems risky, but assuming that it is there is still the issues of visible exposure and the simple economics of keeping an object airborne.
Our offensive technology has certain limitations that cause it to be little better than that of the bipeds. If we can negate the biped weaponry than that would produce a huge proportional advantage. If defensive technologies are practical then there really is no reason to disregard them. And heavy mass-based armour is of limited value against high-energy weaponry and the reduction of mobility will likely cause as much damage as the armour prevents. As for tolerating large losses amongst the drones, we have no experiences with large losses of drones, it may negatively affect our mind, and more obviously, a drone represents a significant quantity of useful material, removing an unnecessarily large quantity of them from our processing centres without an efficient means of returning them is expensive and leaving them undefended in the environment risks them being salvaged by the bipeds which in turn risks duplication of our resources and countermeasures specific to our own nature.
The bipeds possess dominance of the region. Their ability to coordinate allows them to effectively search for us and direct superior resources towards any engagement with us. Remove their coordination and our superior technology will overwhelm their newly unsupported elements. Coordination is the only effective weapon that the bipeds possess...
Drones require significant resources to maintain, we should ensure that we possess the resources to do so prior to acquiring the drones themselves. Increasing our food production by two orders of magnitude and duplicating the reactor several times should suffice for the current phase of our development.
a) Drones lost in battle doesn't require food
b) Our food production is steadily rising
c) We have a stasis chamber, that allow us to save food
Drones require food to develop and maintain, and dormancy would only mitigate those requirements to a limited degree, and likely require additional power requirements to offset this. And while our capacity to produce effective drones exceeds that of the bipeds, two queens will not overwhelm an established planetary colony. To simply overwhelm them with numbers we would need at least dozens of war queens and a production capacity orders of magnitude beyond what we possess.
Hive defences should be primarily evasive. Antidetection measures, limited activity, sensory data of the region, and the ability to swiftly evacuate the site would be preferable fortifications.
a) Limited activity is... limited
b) Evacuation will lead to the loss of resources and provide humans with a nice loot
c) As long as humans don't use heavy orbital bombardment (and I doubt that they can do it) our base is very hard to damage, yet alone destroy or take.
d) If we will create evacuation routes for the queens and most useful drones (that's much easier than full evacuation ) we may sacrifice the base and let humans take it ( and lose a lot of soldiers)
Limiting our activity is of no consequence if it does not inhibit our total achievements, which are more of a function of activity than time. If we are judicious in our activities, avoiding risks of exposure, than we will achieve far more without losing anything appreciable.
If we prepare for evacuation, and are forewarned of its need, then we can easily relocate all of our resources and completely remove any evidence that the location was occupied by us specifically. Such costs would only be a consideration if our preparations for evacuation were insufficient because we had instead decided to fortify a static location.
Sustained ground-based bombardment would be quite capable of killing the plants on the surface, and destroying any surface defences. Leaving us with no way to ward off a protracted extermination campaign against our colony.
In short we should use our main advantage : we don't really care about the losses. My strategy is Overproduce humans, while focusing research on biology, food production, energy generation, industry and land to space weapons
Better is an enemy of good, our current tech level is good enough to win a war with bipeds.
We most certainly do care about losses, losing the queen is unacceptable, losing a drone with valuable qualities is troubling, and each drone represents an appreciable investment.
Our strategy should be to spread into uncontested territory, remain undiscovered, and attain an ability to combat the bipeds with indisputable superiority. We should focus on technological advancement, specifically in the fields of sensory apparatus, automated systems, reproducing our reactor, and space propulsion, and discreet expansion, specifically establishing hives in remote locations where the bipeds will not locate them and they are free to expand further...