SAM and AA missiles. The system of variable coefficients of proportional navigation?

According to the latest update we have:

SAM and AA missiles. The system of variable coefficients of proportional navigation is implemented. Only for MIM-104 Patriot so far. WIP.

Can someone explain a bit more this? Does this mean that ED is now implementing a more clever system of PN and missiles will stop pulling 40G off the rail in the initial part of the interception? Is that this?

Does this mean that ED is now implementing a more clever system of PN and missiles will stop pulling 40G off the rail in the initial part of the interception? Is that this?

that would be my understanding of this, yes!

(i don't have any more information than you though.)

It will generate more confusion actually, since the number of 'APN' implementations are uh ... a lot :D

The gist of APN vs PN is that PN relies purely on LOS rate, while APN also adds an estimate of target acceleration ... I believe perpendicular to the LOS. 'TPN' (true proportional navigation) also adds Vc (closure) into consideration and this has been used on the AIM-7 since almost the beginning ... before the models that we're familiar with. Therefore, you want T/APN combined for best performance typically, but using those instead of pure PN doesn't conserve energy.

Variable PN coefficients shouldn't have much to do with this unless there's a mix-up in terminology - basically the idea here is to vary the common thread among all these, the 'N' ... which basically multiplies the target LOS rate by N and commands the missile to turn at that rate. If you could theoretically drop N to say, 1.1, you would still track the target correctly but the missile would be pulling far less g to do so. And along range, this preserves energy. With slightly more complex algorithms you could drop N further or even delay the maneuver until some form of analysis (ie. computing average motion over some amount of time proportional to the TTG) ... etc.

Please invite Chizh to give a brief technical explanation :D

This is a change from proportional navigation to augmented proportional navigation. A Google search will bring up a number of PDFs on the subject, would probably be less time and confusion than having me try and explain it :)

Odd, it won't load for me. Please ask, if you don't mind, if he is manipulating N based on distance to target. That's what we're looking for ideally. APN isn't really a solution to the worst of problems that we experience in-game with relation to missiles. A welcome change, but not a solution to the energy preservation issue. Might make some missiles a little harder to dodge in terminal maneuvers though.

Thank you, that explains a bunch of stuff, and I was somewhat wrong, but not entirely.

Here they call 'classical PN' what I call 'TPN', which is fine. APN adds the gain variability based on perpendicular target maneuver. I you read the paper, it still exhibits the basic problem that I'd basically like to see gone. Meaning, it still leaves the missile vulnerable to high-g maneuvers when not necessary.

The way AIM-7 handles this is to basically dump N low until within a certain range, among other things. To give you a bit more of an idea:

If you launch with good ASE, maneuvering may even be eliminated during acceleration based on target behavior (prefer to get fast first, especially true in loft).

The missile won't maneuver much until inside 4.5nm of the target, where N is returned to it's full value.

Likewise, lofting requires more specific changes in these rules, namely preventing the missile from the sudden pitch-down experienced currently.

Finally ... it would be nice to also see OCT (optimal control theory, based on zero-effort miss, though there are hints of this in the paper you gave me) ... by manipulating the inputs to the algo you can reduce the functionality from full OCT with/without restrictions to pure PN, thus accomplishing two things:

1. You can differentiate missiles based on algo capability

2. You can degrade guidance via ECM ... as certain values become unavailable, the weapon system must replace them with guesses, so you can degreade from OCT to APN to TPN to PN. Essentially, a fully degraded missile would act exactly as it acts today. Dangerous close in, but quite vulnerable to maneuver defeat at range (this is about maneuver only, never mind countermeasures for now).

Alternatively, since T/APN use Vc, you could pump a very yummy 'closure' signal from the ECM that would make the missile go somewhere where there target isn't ... and at the same time, it isn't exactly easy to tell that it's ECM.

Anyway, OCT is one step further from the APN described in this PDF, and it looks sort of like this:

A=(N*ZEM/t^2go) + (1/2)Natn

Edit: Oh pfft... I guess the above is close to the MGS in that same paper. So yes, that would be a great thing to implement :)

How bout in plain english for those of us who find equations hard to follow :smilewink:

Is this going to end up giving the missiles some more energy to play with than with the current modeling?

Im assuming that if the testing for the Patriot is successful it will be carried over to all the missiles in the game including player controlled air-to-air missiles?

It might make them a bit harder to evade when energy is high. There might be some energy preservation in there, but I don't believe it will be significant. I could be wrong ... but then again, real missiles using all forms of PN have additional energy-preserving measures.

How bout in plain english for those of us who find equations hard to follow :smilewink:

 

Is this going to end up giving the missiles some more energy to play with than with the current modeling?

 

Im assuming that if the testing for the Patriot is successful it will be carried over to all the missiles in the game including player controlled air-to-air missiles?

I too would like to see some more technical details on their missile guidance implementation at some point.

As GGTharos said this change is going to have some impact but there are challenges with these more advanced guidance techniques that mean they aren't necessarily the end all be all. It seems that parameter scheduling, in particular scheduling of N based off a TTG estimate (or just based off the TTG passed to the missile perhaps?), is the more likely improvement that preserves missile energy during long range shots.

Anytime your guidance law uses more target state estimates you typically will become more susceptible to noise and ECM. And there are technical challenges associated with building estimators for these states that ultimately impose limitations on how well the states can be estimated in the first place.

Anytime your guidance law uses more target state estimates you typically will become more susceptible to noise and ECM. And there are technical challenges associated with building estimators for these states that ultimately impose limitations on how well the states can be estimated in the first place.

Having a somewhat accurate state space model is always better than having nothing. :)

Hope for the future of more interesting ECM!

thx ggtharos,

Anytime your guidance law uses more target state estimates you typically will become more susceptible to noise and ECM.

This.

Taking account that ECM is not properly implemented and probably will not. I realize the current missile situation is probably the best option, with the exception of the Mirage missiles in my opinion over performance in comparison with the mentioned.

Hi @NineLine  ! I just checked the .lua file and nothing has changed since 2018? Like, only the Patriot has a variable coefficient depending on the distance between the mssile and the target, no other SAM has it. It has been abandonned like that?