|
Post by alias72 on Apr 16, 2017 18:04:29 GMT
I cannot hit the broadside of a barn with a guided missile.
And it is wearing INCREDIBLY thin.
I have requirements for 2 different controllers.
1. is optimized for a low DV KKV 2. is for a guided missile
the problems I have are:
1. The KKV runs out of DV before hitting the target. The KKV trails the target instead of leading it. The KKV burns continuously at full thrust instead of throttling to match acceleration. in an earlier setup the KKV would tumble while under constant thrust wasting propellant. terminal phase executes too late in the engagement. (also all of the control math assumes a standard rocket when what I wanted was a translational rocket).
2. The missile lags the target instead of leading. The missile runs out of DV early even though that should not be the case for the given intercept. The missile uses full thrust instead of throttling. The missile may lose target lock and drift past the enemy.
Halp me plz.
|
|
|
Post by AdmiralObvious on Apr 16, 2017 18:35:47 GMT
I don't have my PC available, how low DeltaV are we talking about? Mostly because if there's so little, and the rocket has too high acceleration, the AI won't know how to turn off the rocket.
Also, if your guidance is falling short of the target, I suggest using Deviated Pursuit first in the boost stage, that way the missile is already leading ahead of the target, so once you switch to APN, the missile should in theory not have to turn as much for final intercept.
|
|
|
Post by alias72 on Apr 16, 2017 18:52:16 GMT
200 m/s on the KKV. max acceleration 1.9 g's If you want I can give a summary of my requirements and ideal design for this.
580 m/s on the missile. It is basically just a boosted KKV and I may want to do something different later but the KKV HAS to work and it needs to be versatile.
|
|
|
Post by theholyinquisition on Apr 17, 2017 0:09:14 GMT
200 m/s on the KKV. max acceleration 1.9 g's If you want I can give a summary of my requirements and ideal design for this. 580 m/s on the missile. It is basically just a boosted KKV and I may want to do something different later but the KKV HAS to work and it needs to be versatile. Pretty sure that a decent gun should be able to deliver that much delta v. What use do you have in mind?
|
|
|
Post by alias72 on Apr 17, 2017 0:44:17 GMT
I have a cannon that fires a bullet at 2000 m/s. I want to make the bullets semi-guided to compensate for the low rate of fire. Low ROF allows me to reduce the power draw on the weapons and the mass of ammo required. Guided sub munitions are also invaluable in forcing the enemy to dodge. If the enemy does not wish to be hit by a wave of kinetics fired by my ship they must invest 200 m/s of dV. To do this they must reorient themselves such that they expose their broadside to the incoming fire and any laser weaponry I have available. This also puts them in a less than idea position to fire their own weapons at me and allows me to dictate the engagement. Finally while poorly orientated they will not be able to increase the rate of closure so I can use that to dictate my effective muzzle velocity vs theirs.
I wanted a bullet with translational guidance. It would match the acceleration of the target vessel. This was not possible with the existing control laws so I am using a single engine with a gimbal. The bullet needs to lead the target, that is all.
200 m/s 1.5 g min acceleration 25 kg tungsten carbide penetrator
Given a KKV with:
Muzzle Velocity: 2 km/s
dV: 200 m/s
min Acc: 1.5 g's
-no slew time in roll or pitch
-no dV loss to slew or lack of deadband
For a Target with the following:
-Pure Lateral Thrust
-Constant Thrust
-no slew time to dodge
The minimum guaranteed hit range is:
Acc Burn T Range
1.5 13.6s 27.21km
1.25 16.3s 32.65km
1 20.4s 40.82km
0.75 27.2s 54.42km
0.5 40.8s 81.63km
0.25 81.6s 163.3km
In practice the maximum effective range should be larger than this as:
-enemy slew will take nonzero time
-enemy may not thrust continuously
Against targets with excessive Acc:
-dodge still must be performed
-dodge in terminal phase at excessive thrust or for greater than 200 m/s
|
|
|
Post by AdmiralObvious on Apr 17, 2017 1:31:13 GMT
The gun will automatically "lead" for you in theory, I'd suggest using a single low acceleration, higher deltaV engine with about 15 degrees of gimbal, that way, assuming you use APN, or Pure navigation with a (hopefully high) damping factor, the missile/shot should at least try to orient towards the target properly.
|
|
|
Post by dwwolf on Apr 17, 2017 9:00:27 GMT
Yeah. That.
Guidance from missiles launched in an instance can be wonky.
You also have to keep in mind in what the cannon is firing at and what the missile guidance wants to impact ( the hottest target ).
Depending on your thruster they will always be on. Fluorine based missiles are renowned for this.
Also, im not sure if missile guidance understands the fact that its launched with a predictive intercept vector.
|
|
|
Post by alias72 on Apr 17, 2017 16:55:17 GMT
The missile runs on a hypergolic mix of hydrogen peroxide and RP-1.
I have implemented some of the suggestions here and the results are significantly better. There are still issues:
The missile range is 50km. If launched from this range it behaves optimally. If launched within 25 km It skips the boost phase. It then either does nothing or waits for terminal. Part of the issue is I want my missile to perform in a hybrid fashion. I want the bus to accelerate to target and then "deploy" (though in this case switch guidance law) the kinetic impactors. These impactors have their own guidance.
The KKV works much better though still suffers from trailing the enemy. This really shouldn't be the case given the dV but it appears that the missile guidance in the midcourse phase doesn't keep up with the enemies movement. This causes the missile to lag behind such that it has to compensate heavily for the position offset in the terminal phase.
|
|
|
Post by Enderminion on Apr 17, 2017 23:32:56 GMT
I have had KKVs phase through the enemy, their was even 132cm/s of sloped armour for them to pass through, why I think a neutrino would do more damage
|
|
|
Post by dwwolf on Apr 18, 2017 10:20:04 GMT
What ive noticed about gun launched kkv ( mine has a .7km/s dV with a 2.8km/s start) is that they tend to perform quite decently when the firing vector coincides largely with the targets movement vector.
The closer you get to a "T" shot the worse the performance starts to get. What also effes up in instance gun launched kkvs is gun target and kkv target selection....the KKVs IR based targetting has its own ideas about what it wants to shoot.
Side ways launched missiles ( ie from normal missile launchers also tend to confuse the guidance ).
|
|
|
Post by ketatrypt on Apr 23, 2017 14:50:59 GMT
Been playing with gun launched kkv's, and I have been having all these issues - KKV's spiraling out of control, wonky guidance, KKV's phasing thru the target, etc.
The only thing that I have sorted so far is gimbaling - a little is a lot. My current KKV has a 0.1% gimbal, and its more then enough control. (2 sec turnabout time). Wider gimbal angles result in the KKV tumbling and wasting all its fuel.
Another bigger issue that I have been having is getting the kkv to properly track the target. They seem to have no ability to lead a ship that is under constant acceleration, and, rather, try to just follow the target, wasting all the fuel in the process, rather then plotting a proper intercept.
When they do work, they are quite amazing given how my current iteration doesn't even have any sort of payload - its just a cheap 6.5kg 8cm diameter (wet weight) nitromethane rocket wrapped in a thin sheet of aluminum, and they will go right thru all the stock ships I've been able to hit so far. Also they seem really accurate - Through the same hole accurate. A bit more scatter on late terminal would be nice.
I really hope it can get working better, because they seem so much more effecient then any of the other weapons systems, as they don't need massive amounts of power or weight. Just a 1.5t turret loaded with 150 kkv's, sipping on less then 5kw of power, able to launch a guided projectile at 3km/sec.
|
|
|
Post by dwwolf on Apr 24, 2017 8:41:54 GMT
I need to get back to this topic.
My latest guidance parameter iteration is as follows.
No boost, no midcourse phase. Terminal guidance set to augmented pure pursuit. 80% mix , accel on.
Jousting is accurate to roughly 3.5x the nominal range of 17km listed for 1000m^2 ships.
|
|
|
Post by dwwolf on Apr 25, 2017 8:33:42 GMT
I switched my propulsion to a gimballed micro decane NTR (~2g accel/3.5kg package/0.7kms dV /2.8kms gun boost) and guidance seems to be lots better.
I suspect uncontrollable engine burn played a role in poor results.
My current SOP is spitting out 20 to 30 KKVs from 2 cannons with ignore range and prediction enabled. And setting group guidance to midcourse ( which doesnt use dV ). Then setting terminal phase guidance at around 50 to 40 km range.
Getting hits on stock gunships and missile schooners in the 200 to 100km range. Not alot at first.....but 3.5kg KKV leave an impression when they hit.
|
|
|
Post by dwwolf on Apr 25, 2017 9:13:59 GMT
And I think I will switch over more missiles to decane NTR just to see if it improves guidance.
|
|
|
Post by ketatrypt on Apr 25, 2017 12:15:47 GMT
I need to get back to this topic. My latest guidance parameter iteration is as follows. No boost, no midcourse phase. Terminal guidance set to augmented pure pursuit. 80% mix , accel on. Jousting is accurate to roughly 3.5x the nominal range of 17km listed for 1000m^2 ships. Do you find the initial boost causes issues? I have a hard time giving that up, seems it gives the projectile a much shorter flight time. I will have to play with more types of engines.. Not really sure how the ignition/throttling properties work with them tho, so its hard to choose a fuel.
|
|