• Announcements

    • Zapata

      Abbreviated rules   07/28/2017

      Underdawg did an excellent job of explaining the rules.  Here's the simplified version: Don't insinuate Pedo.  Warning and or timeout for a first offense.  PermaFlick for any subsequent offenses Don't out members.  See above for penalties.  Caveat:  if you have ever used your own real name or personal information here on the forums since, like, ever - it doesn't count and you are fair game. If you see spam posts, report it to the mods.  We do not hang out in every thread 24/7 If you see any of the above, report it to the mods by hitting the Report button in the offending post.   We do not take action for foul language, off-subject content, or abusive behavior unless it escalates to persistent stalking.  There may be times that we might warn someone or flick someone for something particularly egregious.  There is no standard, we will know it when we see it.  If you continually report things that do not fall into rules #1 or 2 above, you may very well get a timeout yourself for annoying the Mods with repeated whining.  Use your best judgement. Warnings, timeouts, suspensions and flicks are arbitrary and capricious.  Deal with it.  Welcome to anarchy.   If you are a newbie, there are unwritten rules to adhere to.  They will be explained to you soon enough.  
paddle_master

Applications with wind estimation

Recommended Posts

Dear sailors,

The world of computer science needs your help!

Within the framework of my master thesis I am developing a solution for wind estimation. The solution will use GPS track and boat’s polars as input in order to determine the direction and speed of the wind.

Right now I am conducting a market analysis for existing solutions with similar context.

Do you use any application software (smartphone apps, web apps, pc/mac software) with wind estimation functionality?

Such software may guess maneuver types like tacks/jibes and give feedback about your sailing performance without the need for wind measurement instruments and manual wind data input.

It would already help me if you just right the name of the application you are using.

 

Any additional feedback about the application usage will be much appreciated.

 

I am specifically interested in the following questions:

·      How accurate is the estimation?

·      What do you have to do/how do you have to sail to receive a correct estimation? (e.g. perform specific maneuvers or sail in 3-4 different points of sail)

·      What kind of data is estimated? (e.g. wind speed and direction, sea state, wind gusts, estimation confidence, change of wind direction and wind speed)

·      How does the application react on wind changes?

·      What does the application show if it doesn’t have enough data collected to make a confident wind estimation, for example when the application is starting to record your track?

·      What is the main purpose of the app?

Do not hesitate if you want share any further experience and impressions. Feel also free to ignore the questions above and leave a feedback in free style :)

  

Cheers

 

paddle_master

Share this post


Link to post
Share on other sites

True wind from the polars? Where are you going to get values for current, wave effects, planing, hull and sail conditions, wind gradients...? Too many small errors will destroy the calculation.

Plus, only landlubbers care about true wind, no? I cannot recall ever wanting that info from the boat. From GRIB's sure. But not the boat.

Boat instruments don't really compute true wind, even though the ever optimistic marine software people claim so. It is some kind of water relative true wind, not the weather bureau's geographic true wind, no?

But carry on, I'm just freestyling...

Share this post


Link to post
Share on other sites

Are you trying to solve a problem, or trying to develop a product, e.g. a marketing exercise? 

the calculation of the wind relative to the boat & water is what is typically referred to as "true", while the calculation relative to the earth is referred to as "ground" 

Weather report shows the "ground" wind. 

Typically a well setup system (e.g. H5000 from B&G)  will measure the relative velocity and angle of the wind at the mast head. 

Correct for known offsets of angle, (zero of installation) 

correct for any measured mast rotation  (requires a sensor) 

correct for calculated errors due to heel, pitch, roll, upwash, and leeway. (requires a "9 axis" sensor) 

Calculate for the corrected boat speed through the water. (requires a sensor and calibration) 

display an "apparent" and a "true" wind angle. 

Higher end units then correct out the effects of tidal set and show "ground"

a simple analysis of SOG/COG to estimate the wind would be only useful if you were tacking track data from something like Yellow Brick and back calculating the imputed wind direction and speed from the observed open water sailing. 

 

Share this post


Link to post
Share on other sites

I have a pet project I've been working on for a bit that might have some overlap with what you're doing  Called "Mr Gibbs".

I do have a wind vane but it's really meant more for collecting data right now to generate polars, to eventually be used in an algorithm like you're describing.  My current setup has gps, 9 axis accel/gyro/compass on the boat, wind speed/direction, and 9 axis accel/gyro/compass on the wind unit (which I have mounted at the top of the mast).  It uses the pebble watch for display and control, and I'm working a larger format display that will be more akin to the typical dinghy/sportboat tactical compass.

Anyways, after my shameless self promotion, what I intended to say was that it's hard enough just to calculate true wind with all those sensors, let alone without them.  Current, wind shear, trim and tune all have big effects on how a boat reacts to wind even if you have accurate polars (a big challenge on it's own).  Not trying to discourage, obviously I love this stuff and spend way too much of my free time on it, but it also caught me off guard how hard it is to make something that is practically useful.

Share this post


Link to post
Share on other sites

My sailing app, StartLine makes this calculation.  It assumes you are sailing at your target speed and uses that to look in the polars for the wind that corresponds to that speed.  It takes current into account but you have to enter the current.  There are several problems with what you are trying to do.  Mostly sailors want to know the wind so they can see what their target speed is and make sure they are sailing correctly.  But that is the opposite of what you are suggesting.  In the case of StartLine, the goal is to find out the target angle so that the tacking angle can be calculated (including current etc etc).  As the wind builds, boat target speeds tend to flatten out so determining the wind speed will be very inaccurate.  But in the case of StartLine, I don't really want to know the wind speed, just the target angle, which also tends to flatten out so no harm no foul.  I don't know how accurate the measurement of wind speed is, my guess is that it is not very accurate, but good enough for what I am trying to do as we are talking about a very second order error in tacking angle with wind.  Wind speed itself is not displayed, only the target tacking angle which includes both the variations due to wind as well as current. 

I hope this is helpful to your project.  Let me know if you want any more information.

Share this post


Link to post
Share on other sites
22 hours ago, allene222 said:

My sailing app, StartLine makes this calculation.  It assumes you are sailing at your target speed and uses that to look in the polars for the wind that corresponds to that speed.  It takes current into account but you have to enter the current.  There are several problems with what you are trying to do.  Mostly sailors want to know the wind so they can see what their target speed is and make sure they are sailing correctly.  But that is the opposite of what you are suggesting.  In the case of StartLine, the goal is to find out the target angle so that the tacking angle can be calculated (including current etc etc).  As the wind builds, boat target speeds tend to flatten out so determining the wind speed will be very inaccurate.  But in the case of StartLine, I don't really want to know the wind speed, just the target angle, which also tends to flatten out so no harm no foul.  I don't know how accurate the measurement of wind speed is, my guess is that it is not very accurate, but good enough for what I am trying to do as we are talking about a very second order error in tacking angle with wind.  Wind speed itself is not displayed, only the target tacking angle which includes both the variations due to wind as well as current. 

I hope this is helpful to your project.  Let me know if you want any more information.

Thanks for the reference. I have downloaded the app. But as soon as I press "Start" or "Race", the app crashes on my Samsung Galaxy Note 4. 

I will evaluate multiple approaches for wind estimation. Among them neural networks and graph modelling with costs assigned to each edge. There multiple characteristics which may be extracted from GPS and polars. From GPS track it is possible to recognize maneuvers. I am able to analyse maneuvers considering different correlations which I have already collected from data mining. For instance: speed change during jibes is less than during tacks. Polars may give a clue about optimal tacking angles and speed ratio between sailing upwind and downwind. It is not possible to sail against true wind without slowing down. All that information may be aggregated for estimation of true wind angle. With some curve fitting of actual and target polar chart it is possible to guess at least some wind speed just as reference. 

The solution I am developing is not supposed to replace any wind measurement systems with high accuracy. Instead it should be used for race analysis to produce a default wind information for cases where no wind tracking has been made. Until now Raceqs is the best example for that I got. 

Best reagards,
paddle_master

Share this post


Link to post
Share on other sites
10 minutes ago, paddle_master said:

Instead it should be used for race analysis to produce a default wind information for cases where no wind tracking has been made. Until now Raceqs is the best example for that I got. 

Best reagards,
paddle_master

Having studied a few log files where I had SOG/COG/BSP/TWD/TWS data... you may find that there are confounding variables that are not evident in your SOG/COG analysis. it's surprising how often the boat is not at a steady speed at 100% of polar, though in an open water point to point that may be less a factor. 

Share this post


Link to post
Share on other sites
2 hours ago, paddle_master said:

Thanks for the reference. I have downloaded the app. But as soon as I press "Start" or "Race", the app crashes on my Samsung Galaxy Note 4. 

I will evaluate multiple approaches for wind estimation. Among them neural networks and graph modelling with costs assigned to each edge. There multiple characteristics which may be extracted from GPS and polars. From GPS track it is possible to recognize maneuvers. I am able to analyse maneuvers considering different correlations which I have already collected from data mining. For instance: speed change during jibes is less than during tacks. Polars may give a clue about optimal tacking angles and speed ratio between sailing upwind and downwind. It is not possible to sail against true wind without slowing down. All that information may be aggregated for estimation of true wind angle. With some curve fitting of actual and target polar chart it is possible to guess at least some wind speed just as reference. 

The solution I am developing is not supposed to replace any wind measurement systems with high accuracy. Instead it should be used for race analysis to produce a default wind information for cases where no wind tracking has been made. Until now Raceqs is the best example for that I got. 

Best reagards,
paddle_master

I see the crash report.  It is in the text to speech init.  You are the only crash in I think years.  You got five.  It must be something about the Note 4.  I am on my boat now do cannot troubleshoot it until tomorrow.

Share this post


Link to post
Share on other sites
10 hours ago, allene222 said:

I see the crash report.  It is in the text to speech init.  You are the only crash in I think years.  You got five.  It must be something about the Note 4.  I am on my boat now do cannot troubleshoot it until tomorrow.

The problem seems to be a conflict with a Samsung text to speech app.  The first two hits on Google give suggested fixes.

Here is a link.

By the way, the app also keeps track of the history of tacks so you can review them as help in deciding the upcoming tack angle.

Allen

Share this post


Link to post
Share on other sites

I will evaluate multiple approaches for wind estimation

There's a fine line between tilting at windmills and simply squandering one's life, but I believe this undertaking is squarely in the latter category.

High end race boats invest multiple hundreds of thousands of dollars in attempting to deduce true wind speed and direction from apparent wind speed and direction combined with heading and boat speed data. 

In order to analyze sailing performance, data is logged at several hz and extensive effort is made to document the configuration including sails, appendages, crew, driver, sea state, bottom condition and much, much more. The net effect of these concerted efforts, undertaken by paid, trained professionals, is that we get a general sense of things at best. Often, the result of the analysis is that there is no result.

This is the real experience of the very state of the art.

To believe that you're going to reverse engineer true wind speed and direction from GPS track and polars is pure fantasy. To claim that this is a computer science undertaking is ridiculous. To have sold this to an academic advisor suggests that your real skills are in marketing.

Share this post


Link to post
Share on other sites
6 hours ago, Moonduster said:

I will evaluate multiple approaches for wind estimation

There's a fine line between tilting at windmills and simply squandering one's life, but I believe this undertaking is squarely in the latter category.

High end race boats invest multiple hundreds of thousands of dollars in attempting to deduce true wind speed and direction from apparent wind speed and direction combined with heading and boat speed data. 

In order to analyze sailing performance, data is logged at several hz and extensive effort is made to document the configuration including sails, appendages, crew, driver, sea state, bottom condition and much, much more. The net effect of these concerted efforts, undertaken by paid, trained professionals, is that we get a general sense of things at best. Often, the result of the analysis is that there is no result.

This is the real experience of the very state of the art.

To believe that you're going to reverse engineer true wind speed and direction from GPS track and polars is pure fantasy. To claim that this is a computer science undertaking is ridiculous. To have sold this to an academic advisor suggests that your real skills are in marketing.

I think this is harsh.  It depends on the goals and as the OP said, RaceQs does this kind of stuff and it is very useful.  I do this kind of stuff and don't find it particurally useful but I think there is potential if someone worked on it like the OP is proposing.  I am not motivated to do that because our race legs are one tack so there is just no way to get history on a leg.  We don't get wind shifts and there is more wind at the layline so everyone just tries to guess the layline.

 

Share this post


Link to post
Share on other sites

@paddle_master I did a race last evening and thought you might be interested in my tack log.  I am not saying this is useful but I do have users who find it very useful but they do more tacks per leg than I do.  I usually don't do but one tack per leg but on this race we had a port-starboard situation so I did two more tacks than usual and it shows up in the log.  The race was W, L, Reach, W.  Leg 3 was a reach.

The columns are leg, angle, target angle, latitude, longitude, timestamp.  In the app, I don't display latitude and longitude but instead I display heading.

ST    36    --    37.667765    -122.36365    8/15/2017 6:15
1    92    97    37.66531167    -122.3665883    8/15/2017 6:17
Rnd    45    --    37.66829667    -122.3693067    8/15/2017 6:19
2    5    70    37.69367    -122.3448917    8/15/2017 6:40
Rnd    106    --    37.695    -122.341105    8/15/2017 6:43
4    26    98    37.66550167    -122.330595    8/15/2017 6:59
4    99    100    37.656445    -122.3416933    8/15/2017 7:06
4    95    42    37.66432833    -122.3545317    8/15/2017 7:14
4    75    102    37.66376833    -122.35527    8/15/2017 7:14
Rnd    62    --    37.67237167    -122.3698367    8/15/2017 7:23
5    168    61    37.67551333    -122.3685233    8/15/2017 7:25
5    155    100    37.67335833    -122.36954    8/15/2017 7:28

Share this post


Link to post
Share on other sites

 

On 16.8.2017 at 2:28 PM, allene222 said:

The problem seems to be a conflict with a Samsung text to speech app.  The first two hits on Google give suggested fixes.

Here is a link.

By the way, the app also keeps track of the history of tacks so you can review them as help in deciding the upcoming tack angle.

Allen

Yep, it was Samsung text to speech issue. It failed because I uninstalled ChatOn. And ChatOn I needed to uninstall because my online banking app was not able to coexist with that. When software could just work... But after some tricks I set Google TTS as default TTS and your app stopped crashing. Thanks for the reference. I will take a look on it.

19 hours ago, Moonduster said:

I will evaluate multiple approaches for wind estimation

There's a fine line between tilting at windmills and simply squandering one's life, but I believe this undertaking is squarely in the latter category.

High end race boats invest multiple hundreds of thousands of dollars in attempting to deduce true wind speed and direction from apparent wind speed and direction combined with heading and boat speed data. 

In order to analyze sailing performance, data is logged at several hz and extensive effort is made to document the configuration including sails, appendages, crew, driver, sea state, bottom condition and much, much more. The net effect of these concerted efforts, undertaken by paid, trained professionals, is that we get a general sense of things at best. Often, the result of the analysis is that there is no result.

This is the real experience of the very state of the art.

To believe that you're going to reverse engineer true wind speed and direction from GPS track and polars is pure fantasy. To claim that this is a computer science undertaking is ridiculous. To have sold this to an academic advisor suggests that your real skills are in marketing.

Thanks for your opinion. It was a pleasure to read it, especially the last paragraph about my marketing skills. For an IT geek like me it is a compliment.

 

Guys, its not about the decision, whether I am going to tackle that challenge or not. I have already agreed. The idea comes from a big company which supports me with huge amount of sailing tracks and polars for a wide range of boat classes. For me as a student it is a great intellectual exercise and exciting topic for my master thesis, probably the most exciting one in the whole university. Thats why you have no chance to stop me from my goal!
But you have the option to help me to get through through the most boring topic of my master thesis and become mentioned in my acknowledgments!
If you are using apps like RaceQs, TackTastic, TheFastTrack, iRegatta, Sailware, Sailtracker, StartLine or any other app with wind estimation or maneuver classification functionality without wind instruments, I would really appreciate your feedback about that apps regarding the points mentioned in my first post.

Best regards,

paddle_master

Share this post


Link to post
Share on other sites

It's neither harsh, nor a great intellectual exercise. It's complete folly. To believe you can extract TWD from COG, you don't understand anything about the wind triangle or the vector sum that is heading or course.

Stepping back and looking at your choice of thesis, it's rather like having chosen to prove the earth is flat. It's an endeavor that has no merit and can bear no fruit. As an employer, it would be enough to tip a CV in the bin - the analytics aren't much use when connected to no common sense.

 

Share this post


Link to post
Share on other sites
2 hours ago, Moonduster said:

It's neither harsh, nor a great intellectual exercise. It's complete folly. To believe you can extract TWD from COG, you don't understand anything about the wind triangle or the vector sum that is heading or course.

Stepping back and looking at your choice of thesis, it's rather like having chosen to prove the earth is flat. It's an endeavor that has no merit and can bear no fruit. As an employer, it would be enough to tip a CV in the bin - the analytics aren't much use when connected to no common sense.

 

Have you ever used RaceQs?  They just use gps data (and heel) to do what they do.  Moondust, this is why great inventions are done by the young.  The old know they it be done so they don't try.  @paddle_master, just ignore Moondnust. There are many things that can be learned from SOG, COG, and ploars and if you have enough boat history on the course, you can get a pretty good shot at wind speed, particurally if it is in a mid range and not at extremes.

Share this post


Link to post
Share on other sites
On 8/18/2017 at 1:04 AM, allene222 said:

Have you ever used RaceQs?  They just use gps data (and heel) to do what they do.  Moondust, this is why great inventions are done by the young.  The old know they it be done so they don't try.  @paddle_master, just ignore Moondnust. There are many things that can be learned from SOG, COG, and ploars and if you have enough boat history on the course, you can get a pretty good shot at wind speed, particurally if it is in a mid range and not at extremes.

@paddle_master don't ignore Moondust. Opposite, show his post to your appointed supervisor ..well, obviously sanitized 

"The master’s thesis must satisfy academic standards for publications. The thesis must include a clearly specified research objective. The thesis must provide arguments for the choice of methods used to meet the research objective."

There is no way you can get defendable Master Thesis out this.

 

Share this post


Link to post
Share on other sites

 

@allene222Ok, I did now. Nice apps, good idea, well executed. Smart guys. That's why they allow users to manualy overwrite both wind speed and wind direction when they estimates, based on the rule based algorithm, fail. I guess they are slightly better upwind (with frequent tacks) and much less accurate downwind, specifically on long legs ...

Back to Master thesis thing - not the first time when a big company invests in nonsense. 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now