View Full Version : Using TA+hull sonar for triangulation
Hey again,
I've been playing with the 688(i) again and just had a very strange experience vis-a-vis triangulation using two arrays. I picked up a Han class sub pretty strongly on my towed array (3 frequency lines) and so I thought I might look at my other two arrays to see if I could get him on either. Sphere was blank in NB, but I had a line in the lower window in BB. Unfortunately, DW wouldn't let me mark that line no matter how many times I clicked the designate button on and around its bearing. I then looked on hull NB and bingo, I had the second frequency line coming in feintly. I designated it and merged the contacts in TMA.
Here's where it started to get weird. After a few LOBs appeared, I tried to use the bearing cross from the two arrays to triangulate the target, but the solution I got using the crosspoints was vastly different from what my solution using DEMON said it sould be. I made a new TMA leg, waited for more bearings and again, the triangulation was way off from the solution I got by using just the TA and DEMON.
I got to what I thought was in range and fired two fish, but they didn't seem to lock onto anything and eventually ran out of gas. I was frustrated, so I quit the scenario and viewed the replay and you know what I found? The target was actually way further out than my triangulation suggested. He was at around the exact position my solution using just the TA and DEMON had given me, but I'd thought that the triangulated solution would be more accurate.
Any ideas why this happened? I seem to remember having luck with triangulation before, but I think that was most commonly using the sphere and TA, not the hull. Does the hull array give somewhat inaccurate bearings? Also, why was I able to see an intermediate timescale contact on sphere BB, but was unable to mark it for the life of me?
Castout
10-03-09, 10:27 PM
I'm an idiot in making TMA solution but as far as I know the precise triangulation you mentioned is no longer applicable in DW and it's more realistic that way.
In the previous title like Sub command you will get a perfect precise triangulation when you have a contact tracked with two or more sonar arrays.
I'm not sure as to the exact why but my guess is that for one the different positions of the sonar arrays it self would render perfect triangulation almost impossible. And another that sound doesn't travel in a straight line underwater so you are bound to have bearing differences that would make perfect triangulation almost impossible.
TMA is not a science to me but a kind of an art :damn:. That's another way of saying a thoughtful guessing.
Molon Labe
10-03-09, 11:02 PM
Triangulation with the TA isn't nearly as good as triangulation you get with a UUV because the lines are almost parallel. A very small bearing error can lead to a significant change in range on the solution. Does that mean it's worthless? No, far from it. At times it can be quite accurate. But if you have something better to go on (like a solid unambiguous solution with a known contact speed), go with that.
Something else to keep in mind is that the first bearing lines for new contacts/sensors will be off a little because they aren't on the 2-minute interval so that is going to throw a triangulation off, and obviously there is also the issue where the TA bearing gets thrown off during turns.
I do not know how big the bearing inaccuracies are for each sensor. But you can make best use of it if you beam the contact, and stream the towed array as far as possible.Then the bearings of the 2 sensors are as far apart as possible. If you happend to have your bow and tail line pointed towards or away from the contact then triangulation is useless on it. Then there is no seperation from the contacts point of view.
But you can make best use of it if you beam the contact, and stream the towed array as far as possiblethe problem with use of maximum scope, is, in shallow waters, you'll be dragging your TA, running deaf, and forced to operate at higher speeds to compensate for the drag. Under those conditions, you'll be compromising yourself if you want maximum scope. It's best to optimize maximum scope vs speed based on your environment..
goldorak
10-04-09, 12:17 PM
I do not know how big the bearing inaccuracies are for each sensor. But you can make best use of it if you beam the contact, and stream the towed array as far as possible.Then the bearings of the 2 sensors are as far apart as possible. If you happend to have your bow and tail line pointed towards or away from the contact then triangulation is useless on it. Then there is no seperation from the contacts point of view.
Yes, you need maximum separation between sensors. Next you must proceed with a minimum speed of 7 knots so as to have the towed array in line with ownship. The bigger the depth difference is between ownship and the towed array the bigger the error in the tma solution.
Last but not least, to get a precise triangulation mark the contact on the towed array and hull or sphere sonar AT THE SAME TIME whenever possibile.
This single rule will simply your solution as you can't even imagine.
Edit : when using triangulation 6 minutes (4 minutes to wait for the first couple of bearing lines + 2 minutes for the second couple of bearings line) is all you need to fix a solution. No need to go into several legs, one is more than sufficient if the contact isn't maneouvering.
goldorak
10-04-09, 12:20 PM
the problem with use of maximum scope, is, in shallow waters, you'll be dragging your TA, running deaf, and forced to operate at higher speeds to compensate for the drag. Under those conditions, you'll be compromising yourself if you want maximum scope. It's best to optimize maximum scope vs speed based on your environment..
No, even if you extend the towed array to its maximum length, a speed of 7 knots is sufficient to have the towed array in line with ownship.
No problem of having it drag over the ocean floor.
Go less than 7 knots and yes in shallow waters you'll find the towed array being dragged all over the place.
And frankly the tactical speeds of the 688/seawolf/virginia are quite high that going at 7 knots is no problem at all to search for enemy subs.
The much I remember, what you call "triangulation" was called "master contact". :)
I would rather use the word "triangulation" in a 2 vs 1 situation where the bearings of the hunters are shared to narrow down the exact opponent's position...
That said, TMA has in my opinion rightfully the status of "art of TMA", even more than the "art of evasion" (the last one is an art to me only I think).
I once was good at it, but it is so complex and involves so many decisions that you need constant practicing to keep effective in manual TMA. In a contacts rich environment where your classification quality is poor, this is nearly a job for a professionals (to me).
I dont know if you play online or vs AI, but what I can say is that well done TMA goes far behind than the simple determination of the contact's position... rather it allows to feel the pulse of your prey, and in some games I had the impression I could read my opponent's mind... yes really :)
On the other hand, even here, you'll probably not be able to advance much without a good teacher, as it is plainly too difficult to figure all thing out by yourself (except you have a background already). Best would be a teacher "on the field", practical exercises...
On a final note, i SUSPECT 688 i H/K, SC, DW, and the tons of mods, every single installation will behave differently in nearly everything needing a study by itself (that's what finally pissed me off :))
Another thing that just came to mind.
"Tactical speed": my understanding of this term is, the maximum speed allowed before your arrays start to "wash out(?)", meaning the water flowing on the surface of your sub starts disturbing your sonars, adding noise.
When tracking, it is a good thing to stay at "tactical speed", as you'll get more accurate TMA results, being the lines more distant... meaning the absolute error plays a less relevant role...
NFunky, you pose good questions as I can see from some days now...
Reading them, brought me 5-6 years behind.
I looked into my harddisk and found some documentation from back then. I was a member of the Seawolves without command functions, what I did really like was the battles :arrgh!:
Looking back at the documentation, those were fantastic times. We even had some real submariners, even active submarine officers around...
People i remember are Fish (sort of guru), and many others that I don't read in this forum.
I can say that I was lucky in picking an excellent virtual navy back then (Seawolves), it had the complete hierarchy of a navy. Actually head people were rare or nearly unpossible to talk with heheheh, very resembling after all...
If you haven't considered before, you might want to do some "research" on what is around now as virtual navies, it might become some great experience :yep:
I am not really off-topic, because if you want to understand TMA well, and if you're lucky to find some good instructor/instruction around (for example in a virtual navy), this will boost your skill with much less effort (IMHO).
To tell you one (no joke) I was instructed on Sub Command TMA by a real, active, submarine TMA Officer. Very rarely the questions were so detailed to need an answer like "we're going into classified stuff".
Then it's up to the student to adsorb as much as possible, but you seem to have the requisites to me, as you pose the right questions.
I am unable to make any sort of judgement on the virtual navies around now, but I know some people here are members of the one or the other :DL
Another thing that just came to mind.
"Tactical speed": my understanding of this term is, the maximum speed allowed before your arrays start to "wash out(?)", meaning the water flowing on the surface of your sub starts disturbing your sonars, adding noise.
yessir, this has always been the context that I've used the phrase :salute:
although I prefer akula 2 imp, tactical speed(in this context) is one of the more aggressive, yet flexible features I've always been fond of on the 688.
I'd say she has this very edge over the rest of the submarines in DW; nothing else out there can provide what she has to offer...
Sonoboy
10-09-09, 01:44 AM
*steps out of cryo tube* :salute:
No DW2 yet? :cry:
If this inaccurate solution was due to bearing error, I would expect that you would have some crossover points further and closer than the target's actual position. Assuming that both arrays were indeed tracking the same target, I do not know how to explain this predicament. I will have to try triangulation with hull and towed arrays for myself.
Castout
10-09-09, 02:12 AM
*steps out of cryo tube* :salute:
No DW2 yet? :cry:
.......
get DWX it's the long awaited RA mod in its final glorious form!:D
Make sure to apply the patch too
Still very frustrated with how I can see a contact on the sphere broadband but can't mark it until it becomes pretty bright. This has been causing me grief in high contact environments where all four of my TA trackers are being used and I want to track some of the louder (i.e. surface) contacts with sphere. I suppose I could use the hull array for these, but its got such a low washout speed that I'd really rather not.
Anyway, back to using LwAmi 3.09. DWX, while very cool, still has far too many bugs. I don't need more unknowns (bugs) added to the equation in a game where guessing is the primary task.
goldorak
10-25-09, 05:17 PM
Still very frustrated with how I can see a contact on the sphere broadband but can't mark it until it becomes pretty bright. This has been causing me grief in high contact environments where all four of my TA trackers are being used and I want to track some of the louder (i.e. surface) contacts with sphere. I suppose I could use the hull array for these, but its got such a low washout speed that I'd really rather not.
Anyway, back to using LwAmi 3.09. DWX, while very cool, still has far too many bugs. I don't need more unknowns (bugs) added to the equation in a game where guessing is the primary task.
Nfunky you're going the wrong route with this.
Surface contacts are generally tracked by periscope. It gives you bearing, and distance information through the stadimeter station so 2 measurements are all you need for getting a tma. Speed in this case is irrelevant since it is desumed from the 2 positions you marked.
Let your TA trackers track sub contacts, using those for suface contacts such as merchant vessels is a waste.
You spend time for tma whereas with 2 bearings in persicope + stadimeter gives you a full solution.
You're absolutely right goldorak, I don't know why I'm not thinking clearly. I must be having brain overload from all these contacts, god I HATE trying to cover everything in high-contact environments. I guess I was so busy with the sonar station that I never even thought to pop up to PD and take some pictures. On the other hand, I'd picked up some of the traffic pretty far away, like 20 nm which I believe is outside visual range.
It's funny, in the scenarios I'm playing now TMA is no longer really a problem, but SONAR is such a pain in the ass. Trying to keep situational awareness and tag only certain contacts in the mix is a real headache. The main problem is that multiple contacts share similar bearings far too often and, coupled with mirror contacts from ships on my other side, I almost never know which freqency lines go with which class.
Hint hint, nudge nudge:
http://www.commanders-academy.com/forum/showthread.php?t=3961
This program by someone with the username "Heinrich Liebe" used to be on subguru.com, but it seems Bill linked to the wrong file.
OneShot
10-26-09, 02:25 PM
Take note, the latest version of that file can be found in the file repository at the CADC. I've deleted the attachment to the post and instead included a link to the file repository.
Btw. if anybody has the converter files for 3DSMax 3.1 for DW, please upload them to the file repository and send me a pm afterwards. Thanks
To me the correct way is to track by SONAR. Not always you can track by periscope, if there is hostile air around I wouldn't go to the surface certainly. And... It is not to be assumed that surface contacts (especially if hostile) don't change course/speed. If I am not supported by surface vessels or excellent intel I wouldn't go to the surface anyway (ship's reactivity sux).
If you manage a good classification of your contacts you can focus on the hostile/assumed hostile ones for your TMA work (good classification means good classification in the NAV-MAP). What is certainly not hostile goes green, what's unsecure yellow, what's hostile/assumed hostile goes red). Only if you're going to attack you need a good solution on all the rest around your target. Obviously, try to merge your trackers as much as possible.
Your question about the Hull array, I don't know if things changed lately, but the hull array was primarily there to classify, and not to track.
Hey Nexus7, good points about hostile air cover and such. I do think one of the biggest problems with DW is that my scope/masts can't be detected, so I try to play as if they could be as much as possible.
I guess I am having a bit of a problem when it comes to classification, especially in scenarios where there are surface hostiles mixed in with the surface neutrals. I generally try to have a tentative classification and rough TMA before a contact gets within 10 nm if possible when driving a nuke. When driving a Kilo of course, it's whatever range I can get out of the scope and little sonar suite which is often slightly less than 10 nm.
I was playing a scenario whose title escapes me ATM, but it invovled the interception and sinking of a Chinese cargo ship near Panama. The scenario starts me out with very little info other than that there is a cargo ship in the area with some warship escorts, but no general position is stated or marked on the map. There is what seems to be a plethora of contacts at the start and it took me about half an hour just to sort all these out from their mirrors and one another. Because they were basically all between 90 and 200 degrees, they tended to have tracks nearly on top of one another which made classification EXTREMELY difficult. I don't really have time to turn off the filter and go through the whole list for every contact, so, knowing there were no other subsurface vessels out there, I simply classified most of them as unknown surface. The only ones I could pick out right away were the ones that were pinging, but even so, the track were all so close in bearing it was hard to know which TA contacts to merge with the active intercept tracks. So how do you guys deal with situations like this? I was unable to pick out the cargo ship for about two hours, by which point I had traveled at maximum tactical speed to a point SW of my start in hopes of spreading the contacts out a bit without getting too close. However, even with this maneuver many of them stayed stubbornly close on the TA, though they did shift about a bit. It seems there are just some environments where there are simply too many contacts clustered together to make manual sonar operation anything but a nightmare. And there really were only about ten or twelve ships in the scenario, they were just all in the same area.
In RL, I would assume a sonar operator would have more time to pick out and classify each contact as they would not all be appearing simultaniously at random ranges and speeds. A real sub would be in transit to the location of the intercept and so would pick up surface traffic as it came into range in a much more gradual manner. I would also think that in RL I would get at least a rough idea of which direction to look in for the Chinese convoy and more likely an estimated transit path. Oh well, this one is more the scenario designer's fault than DW's.
Basically it's unpossible/unreal to give you any kind of answer on a simulator that i dont' know anymore.
When you insert the object "RL, Real life", then you would need years of school to still raise questions.
My personal battle is against people acting stupid.
Oh well, I've been trying my best NOT to act stupid or ask stupid questions. My knowledge of (ahem) RL naval warfare is basically nonexistent as I have never even had a conversation with anyone from the navy and there aren't many real life accounts of modern naval operations in audiobook format (I am visually impaired). I was simply extrapolating from what I have picked up here and there on this and other forums. However, I do still think that the way most DW scenarios thrust the player into an immediately contact rich environment with almost no intel or link data makes picking out and tracking contacts a much more difficult task than it ought to be. Were it not for the time constraints of this being a game which should be fun (i.e. not have hours of play with nothing going on) I would prefer to start a little further out from the fray so I could classify and track contacts as they are detected, not have to sort out a dozen all at once while having no idea whether one of them might be an immediate threat.
The only turnaround to this is
- have another person doing sonar (multistation)
- turn autocrew ON
- show truth...
Of course it is pretty normal that classification is a relatively difficult task in a sub... I guess we are way limited in terms of available means to classify stuff. I guess in RL the signature is more exact and different than in our NB frequencies...
For this reason, a scenario with tons of stuff around quickly shows the limits of the simulator... (all russian subs, frequency(1)=50.0000 hmmmmmmmmm. I don't know if the MODS changed that).
btw... I didnt mean that you're acting stupid.
vBulletin® v3.8.11, Copyright ©2000-2024, vBulletin Solutions Inc.