As a side note, if the theory of the pico balloon is correct, then the US airforce is capable of radar-identify a 32" diameter mylar balloon and shoot it with a missile from a fighter jet, which seems to me like being exceptionally accurate.
Radar capability is interesting, but I'm curious if small balloons are really that hard of a target to shoot down once you've found it. They are slow-moving, non evasive, and virtually any impact from shrapnel will be a guaranteed kill. So it seems like you wouldn't even have to be that accurate, if the missile was designed to explode at its closest point.
That said, they did miss on the first attempt with the Lake Huron balloon, so I guess it's harder then I'd have thought.
The problem is in targeting. No missile works on computer vision. There are older guided bombs that use "TV guidance" but I don't think that's even close to the same.
It either needs a hot part for an IR seeker to lock on to, or be radar reflective enough that a high power radar beam can reflect off and be picked up by the missile's antenna and pass a certain threshold of reflected power. There's not much heat in the payloads, and balloons probably aren't the best radar reflectors.
Even gun lead calculations require a good radar return.
I've always predicted we would eventually end up with missiles that guide on visible light, but that's terrifying if it isn't nearly perfect.
The F22's publicly claimed radar cross section is "the size of a bumble bee". It would make sense we build missiles that can attempt to approach targeting that.
