Originally Posted by
Lonewolf_50
I am reasonably aware of the LHX failure, and the risk decisions on missions systems integration on Comanche in terms of cockpit loading, but mission systems in general have come a long way since the 1980's. Army is not good at getting out of old paradigms, and never has been. I will say that I believe that the Apache mafia did as much harm as good in the Comanche's development ... but that's all water under the bridge.
I'd say there's a fundamental reality to flying rotorcraft nap of the earth, which is that the pilot flying has a high workload and a no fail task that automation can't cope with yet unless you give it active sensors (trees, wires, ducks and geese). Active sensors come with a freedom of action penalty due to emcon, so that rules out most automation for now, until computer vision gets really good. Maximum allowable time eyes in for a pilot flying at or below 50' is maybe a second or 2 at cruise speeds. Not really an opportunity there to operate anything or assess a sensor picture.
Contrast that with fast jet aircraft at medium level where the pilot can be eyes in for a few more seconds at a time - here automation of flying tasks and sensor fusion has made single pilot ops feasible.
I would say that advances have meant the non flying crew member can be expected to do a lot more: uncrewed teaming brings in a lot of information but we're a long way from that just being neatly dropped as threats on to a map from an EO picture though AI might help get that over the line to warfighting standard in the next decade or so. (Noting that things a helicopter crew care about tend to be harder to find than something a jet monkey cares about).
Tesla's computer vision kit is probably the most advanced real world example and that can't cope with the real world to appropriate safety standards yet, so we can't really expect a niche application for military rotorcraft to be any better tbh. Everyone that has gone one better is essentially relying on an active sensor of some description (typically LIDAR, which at least is relatively low prob of intercept). Active sensor reliance, as mentioned earlier, limits you due to emcon, but also introduces as massive vulnerability for hostile exploitation unless you have a few alternatives. Demoes of uncrewed Blackhawks going from A to B are great, but it's not tactical flying in a warfighting operation with a bad guy vote.