Thanks for explaining your whole argument here.>Word salad of pointless cope, technobabel and strawmaning and alligning the details to whatever your bis is...
Are you paying me for chemistry lectures and what the fuck does ammonia production have to do with drones?So, haber process with coal, how does it work exactly?
Better explain to everyone what kind of technical parameters does a thermal camera need to have to be capable of doing this at a tactically useful distance of, say, 5km and how much does an sample camera with sufficient performance cost:
That question on the contrary on topic because you're trying to play cool with such inventions and you made a huge doodoo with this here suggestion.And you do not need to be perfect, you do not need to recognize the exact shape of a Stryker or Bradley, you merely have to recognize enough of its characteristics to know wit an acceptable margin of error that you have to hit it and go boom. This can come from the thermal profile of muzzle flared or from a rough shape of the hull done via a simplified image recognition, noting overly fancy, no need for a neural net or other fuzzy logic, a simple DAG with weighted edges could likely yield sufficiently good results given enough input from a more advanced "spotter" drone and other pre-loaded
Furthdrmore, not all A.I. models are as bloated as the stuff you might see and do not necessarily require huge amounts of computing power.
What is in the model is what is important.