Intercept angle and hit probability, room for improvement

[Tcao]

In the latest version of CMO, a new feature has been added where the Intercept angle affects the PH (hit probability). Overall, I like this new feature; agile fighter jets can now evade and outmaneuver multiple BVR missile attacks, making WVR dogfights more likely. However, the formula for how Intercept angle affects PoH is not perfect, and I hope the dev team can improve it in the future. One example is a complaint I raised earlier about the AIM-260 attack H-6, where the hit probability for a head-on attack and a side attack on a bulky bomber can differ by a factor of two, which seems unreasonable.
https://www.matrixgames.com/forums/view ... 0&t=399222

Here's an even more unreasonable example: in this scenario where two AMRAAM missiles attack a hovering helicopter from the side and head-on without any evasive maneuvers, you can see that the hit probability is 48% for one and 92% for the other. The entire difference is caused by the Intercept angle. As I suggested in another post, there are areas for improvement here. Perhaps the missile speed/target speed ratio should have more weight in the final hit probability calculation, and the agility check should be moved to the last step of the formula.
Thx!

missile impact angle.zip

(23.04 KiB) Downloaded 25 times

[caelunshun]

Bump because this issue is still present, and it can be pretty annoying.

I've had scenarios where an extremely advanced anti-air missile like the JATM or SM-6 has a <50% hit probability against a low-tech drone traveling at 60kt just because the interception was perpendicular. It makes sense for fighters that are maneuverable, but for anything else this mechanic does not seem realistic.

The helicopter scenario provided in the OP is another great example.

[caelunshun]

Also, I wanted to ask what real-world mechanic the intercept angle is trying to model?

If it's the "notching" tactic for use against Doppler radars, then it should really not have as severe an effect on modern tracking radars which are pulse-Doppler and should be perfectly adept at tracking a notching target, especially at high altitudes where there is no ground clutter.

If it has more to do with the endgame intercept geometry (head-on intercepts are maybe easier to guide than perpendicular ones?) that makes sense, but the effect should then be proportional to the target velocity. Right now, the effect of angle on a 0kt target and a 1000kt target is exactly the same.

[thewood1]

What's your definition of modern? Into the late 90s over Serbia, notching was still being used somewhat effectively against F-16s and F-18s.

[Dimitris]

If it has more to do with the endgame intercept geometry (head-on intercepts are maybe easier to guide than perpendicular ones?) that makes sense, but the effect should then be proportional to the target velocity. Right now, the effect of angle on a 0kt target and a 1000kt target is exactly the same.

That's a very good suggestion.

[caelunshun]

Into the late 90s over Serbia, notching was still being used somewhat effectively against F-16s and F-18s.

I'm curious about this but couldn't find a source that mentioned notching over Serbia. Do you have more details?

What's your definition of modern?

I'm referring more to radars developed with modern digital signal processing technology, which AFAIK wouldn't include the conflict over Serbia as that was still the era of AIM-120B/C. But I would think AIM-260 and to a lesser extent AIM-120D should be capable of defeating or at least minimizing the effect of notching. Unless your radar is Doppler-only (which almost no modern radars are), there isn't a barrier to tracking a target with zero instantaneous radial velocity besides available processing power to do things like SAR or calculating velocity over time rather than based on the Doppler shift.

(Again, this applies especially at high altitudes where there is no ground clutter that needs filtered out. The need to filter out stationary targets like terrain is the primary reason notching works at all.)

Anyway, regardless of whether notching is effective against modern seekers, I'm sure we can agree a helicopter hovering at 0kts is not capable of using such tactics regardless of the radar seeker in use.

[thewood1]

It was in one of the Osprey Combat Aircraft books. I think it was a Mig-29. There were also several notching incidents over Iraq during and after the Gulf War.

[Dimitris]

If it has more to do with the endgame intercept geometry (head-on intercepts are maybe easier to guide than perpendicular ones?) that makes sense, but the effect should then be proportional to the target velocity. Right now, the effect of angle on a 0kt target and a 1000kt target is exactly the same.

That's a very good suggestion.

Revisiting this now.

The key factor here is the "target bearing rate", ie. the rate of change for the bearing-to-target from the weapon seeker's POV (or the guidance platform's POV, if the weapon is command-guided). This bearing rate changes not only with the target's course and speed relative to the observer, but also the distance (generally the smaller the distance the greater the bearing rate, all other factors being equal).

The problem here is that we have no firm public sources on the linkage between the bearing rate and its effect on PH reduction. What should the "maximum penalty" be? What is the bearing rate threshold at which PH really takes a nosedive? Is it a linear or exponential correlation?

The only relevant source I've come across so far is this: https://apps.dtic.mil/sti/pdfs/ADA364883.pdf - but this again doesn't seem to suggest a direct association. Also going through Fleeman's missile design guidelines to find anything of use.

Any pointers or data that would make this clearer are welcome.

[Nikel]

No idea if this is useful or not, but here is a series of videos by Ben Dickinson on Missile guidance fundamentals.

https://www.youtube.com/watch?v=02APHAy ... ui0tKxdQm7

[Dimitris]

A possibly interesting data point: https://twitter.com/AlexLuck9/status/17 ... 5261646194

On the failed MQ-9 engagement: "Hessen" attempt of interception was made difficult by the drone not directly approaching, but flying at less than 500 kph at distance across horizon (the cross range-issue). A fault occurred in fire control guiding SM-2, causing failure.

So apparently a beam-on engagement does suffer substantially, even against a slow, easy target?

[Dimitris]

Anyone ? With info ?

[caelunshun]

Looks like the effect of intercept angle on hit probability was removed in the latest beta, at least based on my limited testing.

[Dimitris]

It was indeed removed, as being too coarse. We are looking at a more refined system.