A robot firing from a stable platform will fire a bullet more accurately than a human firing the same weapon. The robot weapon’s mount is more stable than your puny human arm. You can fit it with wind sensors and laser rangefinders to automatically compensate for drift and drop. Presumably the sighting optics for both the robot and the puny human will have been properly calibrated beforehand. The only question is whether you need a human to point the camera at the target or is image recognition software good enough to identify a target?
More out of curiosity than doubt: Do you have any concrete examples? I’d love to see some systems in action.
The fusing system is probably redundant. However, the guidance and control system isn’t, save for that it will use several different methods for guidance depending on mission mode (GPS, inertial nav, visual guidence, remote telemetry feedback). Cruise missiles, like all aeronautical structures, are weight critical as any excess weight eats away at payload, range, and battery power. At any rate, having a hardened 3.5 oz projectile striking your cruise missile at >3500 fps is going to–and I’m using a technical term here–tear the living shit out of any structure, so regardless of whether it hits critical avionics or firing ordnance it is likely to do critical structural damage, especially if it hits head on.
Stranger
I went looking on the Wired site for examples, but curiously couldn’t find them. I do recall Wired said they had been very problematic in deployment.
I fail to see the relevance, to be honest. Whether someone is shooting at a known distance on a range or a known distance on the battlefield doesn’t really matter from a ballistics perspective, does it? Either they know the exact range, or they don’t.
If a human can correctly get the exact range on the battlefield using a rangefinder, then presumably so can a robot. If they can’t, then presumably neither can a robot. Either way, it would seem they’re both in the same boat.
I’d say the 120mm. It’s far heavier (4kg versus about 43g) and thus less prone to deflection by wind. It also much faster (1,650m/s versus 850 m/s), therefore has a flatter trajectory and less time in flight. The quoted max range is only 4km for the cannon versus 2km for the rifle, which is a smaller difference than I’d expect. However it may be that the penetrator can’t pierce tank armour beyond that range, even if it can still hit the target.
Going back to the OP for a bit, let’s ignore the use of a robot battlefield sniper.
Instead, let’s look at what we could do with a computer-aimed firing system vs a human. For hitting a target at a given range, using the same type of weapon (round, barrel, receiver, etc), a computer aimed rifle will be incredibly accurate. Constrast our robogun to a human benchrest shooter, trying to place the tightest group possible inside a circle at some arbitrary range.
The problems, as I see them, are as follows:
-
You need to identify the target
-
You need to compensate for range and windage
-
You need to aim the weapon precisely
-
Other small factors to be explained.
-
is easy in the context of just shooting at a target (By target I mean paper thing with circles on it). A video camera system with image recognition software could “see” the target, and figure out where it is in relation to the gun system.
Factor 2 is also easy. A laser rangefinder will give a very accurate range reading. Good ammo will be consistent and the ballistic tables are easily consulted to give the exact correct elevation. Windage is a little more difficult. You can have a windspeed sensor at the gun. There are ways to read wind at a distance, though not incredibly reliably. There is a second method which will work, though. You can fire multiple rounds, but you can fire the follower rounds before the first round even hits the target. Just track the bullet with mm-band radar or some other suitable system, and see how much off your calculations are. If the estimated windage was 1m and the bullet has only moved .5m due to downrange wind conditions, the computer can correct and fire again, keeping track of the rounds as they fly and correcting. Many modern artillery systems use this type of fire control.
Requirement 3 is almost self-explanatory. A computer doesn’t breathe, shake, jiggle, have a heartbeat, etc. It can adjust the elevation and windage of the gun by almost infinitely small increments.
re: 4) – Other small issues I can think of are things like air temperature, coriolis effects, target elevation, barrel droop, cant, air pressure, ammo temperature, etc. All of these are more easily adjusted for by a computerized firing system than by a human.
A real life example of this is MBT firing systems. The Leopard or the Abrams can make first round hits at 3000m at nearly a 95% rate, even on moving targets. A human manually aiming the gun with the backup optical sights is far less able to make these hits, even if the traverse and elevation are powered and the stabilization system is active.
I am sure that somewhere out there is a robot capable of much higher accuracy than any human, and of doing it much faster. Even if there is not, the technology is clearly available to do so.
Of course, this doesn’t translate to a field-ready robotic combat system, and probably will require a little human help to find a target. But hitting that target is going to be much easier for a well-designed computerized system than by a human marksman.
ETA: On preview, the longest tank kill is 5.1km by a Brit Challenger against a T-70 IIRC.
Oh, also, the Chey-Tac LRRS shows just how much having a computer calculate the shot for you can improve accuracy, even when the human is still aiming the gun.
A mere 5mph, full valued wind at the mid-range could throw off your shot over 4ft at 1000 yards.
Thanks. That’s more than I expected.