Autonomous drone swarms will increasingly replace traditional, human-based military units in the near future. That's a fact. It's already starting.
Some early morning thoughts on how these swarms will maneuver in operational space.
Drones are currently in the process of being outfitted with insect mobility -- bees to ants to fleas. However, that mobility is of diminished use given the limitations on decision making complexity (beyond what's required mobility).
That decision making limitation will be fixed in the next decade, as inexpensive computing horsepower and bio-mimicry allows us to outfit drones with more complex mammalian behaviors (think rat).
In fact, given that this decision making capacity will become merely a function of inexpensive hardware/software, it will become a throw away feature. You can turn it up or down depending on need without any thought the expense involved.
This implies a pretty efficient combo of dumbed down drones operating as part of a swarm, reacting to stigmergic signalling, and more rodent like behavior when operating as individuals.
In the field, drone swarms will likely be constrained (as has any military unit across history) by logistics. In this case, it would be the acquisition of energy (less so ammunition).
Autonomous drone swarms will operate independently for months to years. That means:
- Energy efficient maneuver/design. Long loiter for flying drones, etc..
- Ability to hibernate/reactivate.
- Use of natural processes for long distance travel (wind, tides, currents, etc.). Particularly, for very small drones and sea based drones.
Another interesting option is the ability of drone swarms to re-energize themselves through the following:
- Naturally available energy sources. Solar energy. Wind energy (think wind tunnels in urban canyons). The energy acquisition tech will be built into the drones.
- Parasitic acquisition. Human energy sources by attaching themselves to powerlines or sucking gas tanks dry (gasoline is a good energy source for MEM turbines).
- Forward operating bases. Another option is a division of drone labor where a subset of drones acquire and store energy in locations for "soldier" drones to use. These logistical drones could acquire much more complex forms of energy: bio-mass conversion is a good, plentiful option.
I've constructed some pretty fascinating operational scenarios using drones. Probably better as fiction than reality based.
NOTE: You should understand that the evolution of warfare drives social organization (which makes it pretty hilarious that our academic institutions don't study it as a discipline unto itself). The advent of drone swarms will change how our social system is organized. Hint: it's VERY anti-democratic and very similar to feudal models of violence.