SUBSIM Radio Room Forums

SUBSIM Radio Room Forums (https://www.subsim.com/radioroom/index.php)
-   Wolfpack (https://www.subsim.com/radioroom/forumdisplay.php?f=277)
-   -   4-Bearings Technique Tool (https://www.subsim.com/radioroom/showthread.php?t=240711)

Thecoroner 04-14-2019 08:12 AM

4-Bearings Technique Tool
 
1 Attachment(s)
Hello Submariners,

I made a 4 bearings technique helper!

Give your initial course, the time interval you will use and 3 bearings,
It will give you the course of the target.

Then give it your course and the distance travelled before the 4th bearing.
Enter the 4th bearing and it will give you the solution.

You can find it with more infos at :
https://drive.google.com/drive/folde...NLFmWnrsztrf4I

And here's a geogebra construction, I made, to test the correctness of the tool :
https://www.geogebra.org/graphing/ab7wekqx

Let me know what you think.

ljqcn101 04-14-2019 10:51 AM

Good to see more implementations. :Kaleun_Cheers:

I've also made one for more general TMA scenarios, without restrictions on time intervals, number of bearing inputs, or ownship movements. You can check the source code and formulas to see if it's useful to you. (http://www.subsim.com/radioroom/showthread.php?t=240508)

If you're interested you can also check out the tool for Dangerous Waters, which performs line detection on any given image and draw TMA solutions on it, like this:

(green line is the solution)
http://www.subsim.com/radioroom/pict...ictureid=10291

Thecoroner 04-14-2019 12:05 PM

Hi,
I've seen your tool that's nice.
I'm interested in the quad prog optimisation problem.
Especially, in a first place, trying to find the solution going in straight line at constant speed.

I'd like it to be really simple to use. Enter as few data as possible as gathered in the game and get the true course solution.
I should add a bit more info to be able to redraw target true course in the game as easily as possible.

Anyway, it's not perfect and it's hard to get good solutions in game. Hard to be precise with the hydrophone, but at least with the course and rough estimation on distance you can plan on where to go next to get a periscope solution at a good angle.

I'll git my code and check the optimisation problem when I got the time (I shouldn't be doing this :P)

I like the geogebra construction 'cause it's really accurate. You can move the angles to check your solution for distances and courses.

Pisces 04-14-2019 02:23 PM

@Thecoroner: You can't get a solution from 4 lines (or more) by going constant speed in a straight line. You either need to make a course change or a speed change at some point, or both.

Captain_AJ 04-14-2019 04:40 PM

Awesome work
 
There are so many methods for hydro hunting,, I have discovered a few that are less time consuming and are somewhat accurate. the trick is to train your ear for pounding water propeller sounds and learning distance by the sound. I recommend trying a few that are easy to remember and easy to plot in the map& and less time consuming, and will not expose your boat .. these are the methods I use

Bstankos

https://youtu.be/HFhJn8dZI9o

https://youtu.be/hM4c6xqpgic

Thecoroner 04-15-2019 04:49 AM

Quote:

Originally Posted by Pisces (Post 2603264)
@Thecoroner: You can't get a solution from 4 lines (or more) by going constant speed in a straight line. You either need to make a course change or a speed change at some point, or both.

Yeah, I know. But you can still get infos and estimations but the solution won't be unique unless you take a turn or else. As with 3 bearings at stop, you can get the course but not a unique solution.

Thecoroner 04-15-2019 05:08 AM

Quote:

Originally Posted by Captain_AJ (Post 2603286)
There are so many methods for hydro hunting,,

Yeah, you can quite quickly tell if the ship is going towards you or else. I guess only by playing with the gain you could get a range estimation (I can't ^^)
I bet the best can even roughly id the ship like that. (telling if it's a merchant or warship for example).

But it's really hard to pick a point in a convoy and be sure to follow exactly that one to take bearings marks.

I'll check your videos. I watched a lot by stoianm if i recall correctly like 10 years ago. I found some pdf explaining how to 4B running at constant speed (and taking turns) too. I'll try to add that in the tool some day.

Thanks for feed back :-)

ljqcn101 04-15-2019 05:19 AM

You can open up an audio recorder and watch the waveform. But that's cheating anyways.

I think there may also be a way to estimate the speed by counting propeller turn per minute and then multiply by a constant number (something like TPK in Dangerous Waters). But we got over 160 types of merchants and those sound signatures may all be different.

ljqcn101 04-15-2019 05:27 AM

Quote:

Originally Posted by Thecoroner (Post 2603246)
I'm interested in the quad prog optimisation problem.

That's good if you want to add constraints. You can check AdamMil's website to see his description: http://www.adammil.net/blog/v103_Per..._Analysis.html

Since I don't have problems like ghost solutions on the wrong side of the observer, and adding course/speed constrains is more of a DW thing, I just stick to BFGS algorithm for optimisation.

ljqcn101 04-15-2019 05:39 AM

Quote:

Originally Posted by Thecoroner (Post 2603365)
Yeah, I know. But you can still get infos and estimations but the solution won't be unique unless you take a turn or else. As with 3 bearings at stop, you can get the course but not a unique solution.

Yes, but you can get a bunch of solutions starting from 1000m to 5000m for example. (This is the range I set multiple starting point for optimisation. There could be more solutions.) Course solutions can be in a certain range though. But if bearing lines have intersections, you could get course solutions in totally different directions.

Code:

time t1 = 0 (sec)
ownship heading at t1 (deg): 0

target bearing at t1 (deg): 90

ownship straight direction from last position = 0 (deg)
ownship straight distance from last position = 0 (meter)
*******************************
time t2 (sec): 120

ownship heading at t2: 0

target bearing at t2 (deg): 85

ownship straight direction from last position (deg): 0

ownship straight distance from last position (meter): 300

*******************************
time t3 (sec): 240

ownship heading at t3: 0

target bearing at t3 (deg): 80

ownship straight direction from last position (deg): 0

ownship straight distance from last position (meter): 300


target course: 358.88deg, speed: 6.26725knots, distance: 999.996m
target course: 358.16deg, speed: 7.67643knots, distance: 2000m
target course: 358.17deg, speed: 7.66992knots, distance: 1995.38m
target course: 357.67deg, speed: 9.08642knots, distance: 2999.99m
target course: 357.68deg, speed: 9.07374knots, distance: 2991m
target course: 357.31deg, speed: 10.4969knots, distance: 3999.99m
target course: 357.04deg, speed: 11.9077knots, distance: 4999.99m

probability of target course error within 5deg: 13.6364 %
probability of target course error within 10deg: 23.0909 %
probability of target course error within 20deg: 49.2727 %

probability of target positional error within 75m: 3.81818 %
probability of target positional error within 150m: 13.6364 %
probability of target positional error within 300m: 27.9091 %

*******************************
time t4 (sec): 360

ownship heading at t4: 0

target bearing at t4 (deg): 75

ownship straight direction from last position (deg): 0

ownship straight distance from last position (meter): 300


target course: 358.34deg, speed: 6.26019knots, distance: 1000.58m
target course: 358.37deg, speed: 6.23728knots, distance: 984.243m
target course: 358.35deg, speed: 6.25775knots, distance: 998.846m
target course: 358.36deg, speed: 6.24476knots, distance: 989.58m
target course: 358.39deg, speed: 6.21228knots, distance: 966.41m
target course: 357.3deg, speed: 7.66408knots, distance: 2001.17m
target course: 357.32deg, speed: 7.63115knots, distance: 1977.72m
target course: 357.31deg, speed: 7.64557knots, distance: 1987.99m
target course: 356.57deg, speed: 9.06974knots, distance: 3001.76m
target course: 356.59deg, speed: 9.02861knots, distance: 2972.5m
target course: 356.58deg, speed: 9.0449knots, distance: 2984.09m
target course: 356.04deg, speed: 10.4765knots, distance: 4002.35m
target course: 356.06deg, speed: 10.4307knots, distance: 3969.78m
target course: 356.05deg, speed: 10.4356knots, distance: 3973.29m
target course: 355.64deg, speed: 11.8839knots, distance: 5002.94m
target course: 355.65deg, speed: 11.83knots, distance: 4964.62m
target course: 355.63deg, speed: 11.8987knots, distance: 5013.45m

probability of target course error within 5deg: 24 %
probability of target course error within 10deg: 48 %
probability of target course error within 20deg: 81.3333 %

probability of target positional error within 75m: 4 %
probability of target positional error within 150m: 16 %
probability of target positional error within 300m: 34.6667 %



All times are GMT -5. The time now is 11:18 PM.

Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright 1995- 2019 Subsim