Problem when estimating target position via sonar

[Antba]

Hello!

I am newbie on SH4 and I am having some problems. I like simulators to come as real as possible to the real situations, so I ask everyone to respect this option of mine to choose realistic options even if I have few knowledge about subs. The learning may be harder, but it's much more fun!

I will be direct here to ease the understanding and the further helping. I'll provide as many information as I can, so you guys may use what you think is necessary:

Simulator: SH4 with TMO 1.8 beta (ONLY MOD). Everything in realistic, except for external views enabled

Submarine: Tambor

Date: February 9, 1942

Mission: Sink enemy merchant shipping around Honshu

My Objective: Intercept a merchant detected by the radar at a longe range

Situation: Contact is moving away from me, so I want to get ahead of it, going parallel and maintaining a certain lateral distance, then a turn to stay at a 90º angle with the target to make the shot

Method: Use sonar to get target bearing and approximate distance, write marks on the map about its position with 10 minutes difference from one to another, then estimate its course, speed, and overtake him without visual contact to take position for the shot.

First mark (06:48)



Bearing 324, increasing speed, moving away
Estimated ranges:
-22630
-22470
-21920
-23140
-21230
Final estimated range: 22278




Second mark (06:58)

Bearing 324, constand speed, moving away
Estimated ranges:
-23330
-22130
-22470
-21010
-19370
Final estimated range: 21662




Third mark (07:08)

Bearing 325, moving slow, moving away
Estimated ranges:
-22000
-18940
-19470
-21430
-21700
Final estimated range: 20708




Fourth mark (07:18)

Bearing 325, moving away, moving slow
Estimated ranges:
-17810
-18880
-19260
-20740
-20260
Final estimated range: 19390




As you can see, his trajectory looks strange 'cause it's a curve (nothing against ships doing curves I just think it's not the case).

So I took more data 20 and 30 minutes later, at 07:38 and 07:48

Fifth mark (07:38)

Bearing 326, moving slow, moving away
Estimated ranges:
-15690
-18290
-17030
-15800
-18520
Final estimated range: 17066

Sixth mark (07:48)

Bearing 326, moving slow, moving away
Estimated ranges:
-16120
-16570
-16760
-15910
-15250
Final estimated range: 16122





It looks like this trajectory is really offset. The estimated trajectory in sonar tutorials look so much closer to a straight line than mine, so it's difficult for me to estimate the target's future position with this 'zig-zag'.

And I couldn't demonstrate here, but I've already tried this approach a couple of times and the target always is at positions I wasn't expecting. I suppose it is making some hard turns sometimes, but it's not the problem. My problem is to estimate his present and future position even if it looks to be on a straight line and constant speed. Could the merchant know about me and is changing course/speed everytime? Or am I doing something wrong with the measurements? I just want a more-or-less-straight estimated trajectory.

Thanks!

[TwinStackPete379]

Good work on researching the intercept and attack strategies. Now ur measurements are not off, it's slop from the sonar. my experience was the closer you are to the target the slop cleans up a bit. now you got everything you need, all you do now is "eyeball" the course of the target by making a line down the "average" of ur readings. some of ur marks will be on the targets port, some starboard. but you'll have a really good idea of where's he's going. as you close in, fire off a few more active pings to clean up the data and u'll have it. get visual with the periscope, and make a few last minute adjustments and you'll have steel on the ocean floor no problem

[Pisces]

Your plots do not resemble a target that is weaving. The sonar is just very uncertain about the real range to the target. Look at the variability of the returns you got. The largest and smallest returns are roughly 2k/3k yards apart most of the times. Perhaps you could have plotted the max and min ranges each time. And connect them with a short line. (similar to a stockmarket high-low barchart) Over time that would narrow the possible courses down, as the track needs to fit through all of them. Just plotting the average makes you think the plot is more accurate than it realy is.

[Donkey-Shot]

Hi,

The target isn't zig-zagging, sonar is that inaccurate.
It looks like you're using sonar, both active and passive, on the surface.
Not a verry realistic situation, especially at the speed you're travelling.

The short of it is: your target is moving in a straight line, but sonar won't give a perfect range of bearing. And if you want to be realy realistic: don't use sonar on the surface. It shouldn't work.

I know sonar has some capabilities on the surface, but not anything near as good as in your example. I'm sure others on this forum can tell you a lot more about that sonar capabilities.

[magic452]

Welcome to subsim mate.

To improve the accuracy of the range you must be accurate in the bearing.
What I do is move the bearing left till the green light go off, say at 30°,than go right till the light goes off, say at 40°. I than split the difference at 35° and point at that bearing. Ping at that point and your range should be more accurate, not right on but better.

But as stated above there is slop in the system and at long range such as you are using the best you can hope for is a good estimate of the target course.
This should be close enough to plan an intercept. Once you get closer you can get more accurate ranges and a better plot.

It's a lot of fun to track down a ship using only sonar, sounds like a little practice and you will be doing fine.

Magic

[Wilcke]

You may have a different opinion. But that is a great plot! End around and attack! Well done!

[gutted]

As the others said, you gotta average it out. Draw a straight line through the middle of them, so that they are roughtly equally left and right of your line.

looks to me in your screenie the target is heading due south.

if you are trying to get speed between two marks. dont measure between the points themselves, but where they would be if they were on the line itslf.

normally when i do this, i'll leave a line that crosses through the targets path from the directin i was looking from at the time of the mark. i do this for the two plots im going to get speed from. Later on.. if i adjust the target's course by movin one end of the target's course line, the targets exact location on those "sighting" lines will be correct no matter how much i move his course line.

eh.. hope that made sense.

basically if you have the targets course nailed down pretty accurately on the map, you dont need to range him to place marks on the map anymore. just draw a "sighting" from your ship on the correct bearing to where it crosses the targets path and there he is. So what i do.. is throw down some sighting lines early.. then later on when i get his course nailed down, i'll measure between those two sighting lines to get his speed.

[Antba]

I appreciate all of you guys opinions and tips. It's good to know I wasn't using sonar in a wrong way, although my plot needs some more techniques to be more accurate and useful. I will follow your suggestions.

Curious also that the sonar is less effective on the surface and at high speed. I will simulate this aspect when I get my measurements, as it seems the game doesn't simulate.

Thank you all

[harag]

Hi Antba

I'm also fairly new to tracking ships and found some of the above answers really interesting. I've also read elsewhere not to use ping as much, you never know whos listening. I found the below video VERY useful to find the course and even the speed of a ship by just using the hydrophone and NO pinging. Hope you find it of use.

http://nefelodamon.blip.tv/

And to the guy who did the video, if you read this. THANKS

H

[Armistead]

I usually don't ping because of warships. If they pick it up the ships will zig or possible change course. Also, depending on how long you track, convoys will change position.

I use the same method except passive sonar. Really all I get is a bearing.
As in the war, I come to a complete stop so I can easily figure their direction, make my bearing marks. That gives me a general base course, so I'll start my end around, stopping dead a few times to assure track.
Once I get a visual on a merchant from ahead, I can fine tune for a better track of attack.

It's not easy, because the sonarman and radarman in reality by counts and better stuff to work with could figure get speed heading and base course by sonar or radar and have it quickly for you. Thus most play with contacts on. I alter depending on time.