SAR Plugin

[bcn]

Originally Posted by Gilletarom

In french, here, Leg = Segment. So very long word.


And OpenCPN already use "Segment" for other similar item. So impossible to change the word. Probably, I use "Seg."
Do not forget to insert "Leg" in crowdin file.

In German it would be "Strecke" or better "(Teil)Abschnitt".
Not that short neither...

[Rasbats]

Originally Posted by Gilletarom

Ok. Idea :
So, to help beginners, replace the item " Select SAR datum point " by "Selected waypoint as SAR datum point".

But it does not need to be a waypoint.

Mike

[rgleason]

Steve, Mike wrestles alligators with some skill.
The key question for me, is everybody happy about the methods and reliability of data entry now?

[Ops.Navig8r]

Hi gentlemen, As previously discussed I generate all these patterns in my own software so language issues with waypoint naming is not an issue but I personally use terms like "WP-!" and so on however in the Vector Search (VS) "Vector Single' (sometimes called circular or Sector search) after the Datum or CSP I use "Leg 1 Goto" which is long but once zoomed in its OK, but you must remember I send these to other vessels so they need to know what I'm asking them to do. In multi vessel searches the waypoint also contains the boat name or call-sign (up to 6 characters with the waypoint No Start or Goto) so its all about zooming I also track all vessels so I zoom right to see how far they are of track and I want that less then 20 M.
Cheers

[rgleason]

I've tried each of the patterns using the selection from a waypoint, and have several observations .
The plugin menu can be resized nicely to eliminate the graphic.
The spacing, clearances around boxes and size of title text might be made much smaller to save space.
The accuracy of the sector search isn't too good, see where it comes back to the center, has anyone else checked the others?

[Ops.Navig8r]

Hi Rgleason, I didn't download your (VS) vector search or Sector search so I don't know the leg lengths you used but I guess its safe to say the longer the leg the greater this error is likely to be. One easy solution is to only use the CSP (Datum) waypoint for the transitional centre point and only calculate the outer point from that, this will probably reduce any other errors as the length from one to the other is the same and will not increase the error from one to the next accumulating as it goes. Also as these single vessel searches are based on the visibility range of the craft and the (VS) in particular for small targets say a person then the leg length is a serious consideration when planning the type of search. in my software the choice of leg length is based on the visibility, in the case of the Parallel track it is not the leg length but the Space between tracks (less than twice the visibility). So I have a dropdown (VS) 5, 4, 3, 2, 1 & 0.5 nM
The same for the Quadrant Oil Rig search, (ES) 3, 2, 1, 0.5, 0.2, 0.1 nM.
All that said I personally only use these single vessel patterns if we are sure we are close the splash point or searching for small targets and then with small leg lengths under 1 nM, the (PS) would be the exception to this as its likely you are looking for a vessel and not real sure were it is so the increased space will allow greater coverage of an area without compromising the POD too much.
Anyhow I would challenge anyone to steer there vessel around the (VS) pattern and get as close to the waypoints by less than the error you have seen, however I under stand your concern for accuracy, there will always be some error as the world is not a perfect sphere and Longs are not parallel you'll note your start and end point are close as far as lat is concerned but the long is out some metres with the maximum recommended of 5nM the error is about 14.5 M.
Well that's my thought on this and one can sometimes be too pedantic but if you understand the math you must expect some errors even if the resolution of the numbers is 12 decimals, as soon as (PI) is introduced you have some sort of resolution issue let alone the plotter program and chart inherent errors, then throw in your GPS dilution of position. ??????
I produced a 2 pass plot so the error is increased as the total distance is 90nM and the error is about 24.1 M which is about twice the error you S&R plug in produced with a total of 45nM as I would expect.
I have attached a GPS file from my program with 5 nM legs and two passes there is no visible error if you open the file with a text editor you will see a very small error >0.0001 in lat/long of the start and end points so you may get a better result perhaps on how the approach to the calc is done still I think its not too bad (I'll stick with mine....OLO)
Sorry my pattern will display in South Australian waters.
You will also note the one GPX file has two routes with two colours which Mike thought could not be done, I think he will look at that when he is out of the swamp with the other S&R issues.
Best Regards
Steve

[Ops.Navig8r]

Hi RG,
I forgot to mention in my last reply the VS, SS and Oil Rig search are generally not actually performed by following a plotted route or waypoints, we of course need to do this to present the pattern on our plotter. The accepted method is to start at the CSP which ion the case of (VS) can be marked by a drift marker with similar drift characteristics as the target, the first leg is then run in the direction of the drift/tide and run at a set speed, timed by the navigator. As the vessel runs back across the CSP (VS) drifting it should correct its course to pass over the markers new position. So the actual track will distort by leeway and set of the marker and the vessel, therefore chasing the targets drift. The (SS) (Square Single) or Expanding Square and oil rig will do the same but a drift marker is not required as the route does not go back over the datum point but both will distort toward the possible target due to leeway & set. So actually steering to waypoints may reduce the POD for the search.
When I'm set these searches the craft is only given the CSP lat/Long, the first leg length and maybe the bearing, when tracking vessels performing these patterns I expect to see this distortion useless there is no tide or wind effects and I'm tracking power vessels. if your doing this from a yacht then god only knows what your actual track would look like, tacking reaching, running, jibbing with much more leeway and so on. As none of the other pattern end at the start you would never notice errors of a few meters in most cases most can't steer the route that close for too long, I endeavour to get our skippers to steer within 10M and in some seas that is a big ask, but if in poor conditions searching for the person or worse we may employ a vessel space of 50m so if we have 10 vessels in a parallel track not steering very tight there is a high collision risk which is just something else to watchout for while you are supposed to be searching.
Cheers Sorry to throw more wrinkles in to S&R
Steve
Cheers,
Steve.

[rgleason]

Steve the legs were 10 nm I believe. Perhaps it is not that much of an issue, and good that there is some error and diversity, as there's a better probability of covering new ground in the search. Just thought I would bring the situation up for review.

[Rasbats]

Steve...

No problem with your wrinkles. Working from IAMSAR I notice the distinction between CSP and Datum. I am incorporating this and the IAMSAR naming in the code at the moment. For vector searches this means starting the pattern at the CSP not the datum.

For the parallel search (single vessel) I am working on a CSP that is 2 1/2 track spaces separated from the centre point. The search leg orientation is then the search leg course and the pattern can be continued beyond the centre point to allow for drift.

Mike

[rgleason]

Originally Posted by Ops.Navig8r

Hi Rgleason, I didn't download your (VS) vector search or Sector search so I don't know the leg lengths you used but I guess its safe to say the longer the leg the greater this error is likely to be. One easy solution is to only use the CSP (Datum) waypoint for the transitional centre point and only calculate the outer point from that, this will probably reduce any other errors as the length from one to the other is the same and will not increase the error from one to the next accumulating as it goes.

Yes, that might be more accurate but perhaps it is ok as the search in itself is not pin point either.

Also as these single vessel searches are based on the visibility range of the craft and the size of the target..

(VS) Vector Search in particular for small targets say a person then the leg length is a serious consideration when planning the type of search. in my software the choice of leg length is based on the visibility,

(PT) Parallel Search in the case of the Parallel track it is not the leg length but the Space between tracks (less than twice the visibility).

Dropdown
(VS) 5, 4, 3, 2, 1 & 0.5 nM
(QOR) 5, 4, 3, 2, 1 & 0.5 nM

I personally only use these single vessel patterns if we are sure we are close the splash point or searching for small targets and then with small leg lengths under 1 nM, the (PS) Parallel Search would be the exception to this as its likely you are looking for a vessel and not real sure where it is so the increased space will allow greater coverage of an area without compromising the POD (Point of Departure?) too much.

Anyhow I would challenge anyone to steer their vessel around the (VS) pattern and get as close to the waypoints by less than the error you have seen..there will always be some error as the world is not a perfect sphere and Longs are not parallel you'll note your start and end point are close as far as lat is concerned but the long is out some metres with the maximum recommended of 5nM the error is about 14.5 M.

..as soon as (PI) is introduced you have some sort of resolution issue let alone the plotter program and chart inherent errors, then throw in your GPS dilution of position. ??????

Error Comparison to spreadsheet
I produced a 2 pass plot so the error is increased as the total distance is 90nM and the error is about 24.1 M which is about twice the error you S&R plug in produced with a total of 45nM as I would expect.

I have attached a GPS file from my [spreadsheet] program with 5 nM legs and two passes there is no visible error if you open the file with a text editor you will see a very small error >0.0001 in lat/long of the start and end points so you may get a better result perhaps on how the approach to the calc is done still I think its not too bad (I'll stick with mine....OLO)

Sorry my pattern will display in South Australian waters.
You will also note the one GPX file has two routes with two colours which

Mike thought could not be done, I think he will look at that when he is out of the swamp with the other S&R issues.

Your points are good. Perhaps Mike will be able to make it more accurate and do a 2 pass VS.

I downloaded your attached file and see how you made the 2 color, 2 pass plot of the Vector Search. Perhaps that can be done automatically at some point when "2" pass is selected. We'll see what Mike comes up with.

The Opencpn plugins incorporate a wide range of specialized skills and our Plugin Manuals (wiki) are intended to help users learn how to effectively use the plugins. I think that our SAR plugin wiki, which has been improved substantially from the origin version, does need some experienced eyes to review and make comments and improvements.

https://opencpn.org/wiki/dokuwiki/do...ins:safety:sar


Your comments above on selection of search distances based on the target height and visibility, search vessel visibility, sea conditions, weather conditions might be a new section in the wiki.

Also I think the Terminology section needs to be expanded and should include IAMSAR and its link

Also, I think I need to review your comments about how you deal with Set and Drift more carefully, but our wiki does not deal with that in any way other than a brief note at the bottom, and I think there are probably improvements to the wikii that should be made.

If you could look at the manual and make suggestions it would be very much appreciated. I can also make you an editor so you can make changes yourself. Just Private Message me with your email and I will add you to the wiki editors. Perhaps the plugin manual could serve as preliminary instructions for your search team?
Thanks.

[Ops.Navig8r]

Hi Ron, I think, 10 nm I think is way to large a diameter for a (VS) search. that means you must have a good visual range of 5nM else you could miss objects in between legs. These searches except the (PS) are generally for smaller targets and for you guys on vessel planning then the most likely use would be MOB. I that case the Oil Rig would be a good choice especially if following the calculated route. After all if you have a MOB drifting north why would you use a pattern that sends you South? As I said before these patterns usually Drift with W&T.
US Coast Guard have a program you can download, at https://sites.google.com/site/uscgau...d-sar-patterns
Royal Canadian coast guard also have a phone app, some software also uses "Cables" for the legs or radius so you can see the intent is for small distances. I know on board the USCG vessels they us a "Weems & Plath" Leg Calculator for (VS) & (SS) patterns. These can be purchased online from good Map Shops.
I have made work sheets for our on board navigators to use to help keep a record for these "smaller" patterns which I have attach for you to look at. These forms have a Distance/Speed/Time table similar to the Weems Plath calculator. So as I said previously if instructed to perform on of these searches they only get the CSP and the desired leg length (which they may by consultation change due to visibility or sea state) then use the form, no other waypoints, the skipper determines best speed for the target size and sea sate, the navigator keeps an eye on the bearing and time for each leg.
So there is more than one way to kill a cat but you must look at the practical implications of setting hard and fast waypoints for some patterns.
On board as a first responder these patterns are OK but if a search develops to something more than one craft then an S&R agency will take over and you would likely be asked to move on or if utilised (unlikely) do as your directed by them anyhow.
Cheers
Steve.

[Ops.Navig8r]

Answer to a previous post I missed about data entry, now the S&R plugin can be minimised off the chart for easy cursor positioning I could not get the 'Ctrl+S' to place the lat/long in the minimised form, only when Maximised. is it just me?
Steve.

[rgleason]

FYI The 10nm entry was just a number entered, no thought about any of the factors.

[Ops.Navig8r]

No worries about the test leg length you tried, I assume Mike is reading this and he asked about terminology which sometimes varies form source to source, he spoke about the IAMSAR manual, I took a look at it and the thing to keep in mind with it is it is slanted to Aviation search, so the example of the (VS) search has the CSP on the outer edge of the pattern and not the centre, for a marine point of view we would use this where we have located an object relating to our target, then deploy an SRU to more closely search that area from the centre point for more objects, every object is plotted with the location and time as to make more accurate estimates of drift and projection back to the actual splash point or incident. So there may not be the hard and fast rule as to where the Datum or CSP is in relation to the pattern or weather they are one and the same. Usually the SMC decides the pattern and where the SRU's begin. As I eluded to as an organisation we in the main use (CM) Creeping path Multi SRU, searches where we have a number of vessels line abreast effecting a parallel tack similar the (PS) but with multiple SRU's on the one track. I all the documentation I have read I have not found such described which is why I wrote my own software for it. To keep the SRU's in line and easy for them to navigate/steer all searches run to cardinal point for a leg length as required, the CSP is given with the space between SRU's in points or decimals of a minute either lat or long so every vessel then steers by maintaining their minute space in the line, at the end of the leg they all turn line astern to the next leg start then turn again line abreast for that leg. As the math adding or subtraction point of a minute to a lat or long is easy most of our navigators will work out all the turn points on the boat without the base needing to tell them any other information.
STEVE.

[Rasbats]

Originally Posted by Ops.Navig8r

Answer to a previous post I missed about data entry, now the S&R plugin can be minimised off the chart for easy cursor positioning I could not get the 'Ctrl+S' to place the lat/long in the minimised form, only when Maximised. is it just me?
Steve.

I think you will find this with other OpenCPN plugins. You have 'used up' your keystroke when you minimised. The dialog can be resized and it can be 'shunted' out of the chart space. Alternatively the 'Right-Click' menu option can still be used, even when the dialog is minimised.

Mike