Some questions and ideas

[Tommi]

I've got some question and some ideas to propose to you. I don't know if something I'm going to say can't be done by an android app like Talos but I see it like a future enhancement or something to develop. Talos is a beautiful app for a coach (especially the possibility of video sync and the real time remote viewing) but to become a really useful tools for a common rower the app should become able to give a real time feedback also to a technical inexperienced rower... To do this my ideas are:

1) The introduction of an example "golden" acceleration curve (quite hard to do without highly trained athletes to test and without consider many other variables..) or at list the chance to see a comparison between the current stroke and an average curve based on previous strokes. Beyond this the average acceleration curve should be something that can be saved for each rowing session, comparing many of these curves could be a strong technical analysis method to evaluate improvements of the rower's technique.
2) Introduction of a units of measurement for the quality of the stroke, measured at each stroke: this unit could be a pure number derived from the sum of each acceleration values of the stroke. This number will be higher for longer (in time and "angle") or/and for more powerful strokes so it could roughly evaluate the stroke efficiency.
3) Introduction of a units of measurement for the quality of recovery, same technique as the stroke one but these number should be as close to zero as possible (for a recovery without great negative forces on the hull).
4) Talos could produce a time measure of: -how long are stroke and recovery and also which is the ratio of this two values -how much time the rower employs to reach maximum acceleration (this could be a way of evaluating the technical quality of the catch) -how long a rower could keep the boat at an acceleration that is at least the 60/70/80/90...% of the maximum acceleration value that he is capable of.
5) Show acceleration graph in m/s^2 and the values (average and max) of the roll angles.
6) Is there any way to improve the signal to noise ratio of accelerations values? Maybe develop some standard levels (instead of numerical values) to chose from in "not advanced" settings and also create more complex and customizable filters for advanced users.
7) It seems that my Samsung Galaxy SII could not reach decent GPS accuracy, Talos always shows the red indicator in his gps data. This is true always when other app are able to provide 99% gps accuracy (less than 5m error). Is it a bug?

I'm not able to direct develop an android app (I'm going to learn...) and help you to find answers to my questions, but I'm a coach and a rower so I'll provide real word testing and scientific support if needed (I'm studying sport sciences in the University of Pisa, Italy). I'm kindly supported by VVF Tomei Rowing Club and specially by his coach Stefano so we can study and develop Talos Rowing quite well. I'm happy to contribute somehow in Talos Rowing development =). I hope my question will be useful! Thank you all! (Sorry for my poor English)

[Tal Shalif]

Thank you Tommi. As a Talos Rowing developer I find this message wonderful for the following reasons:

1. You may be able to provide 'golden' rowing data from elite rowers
   
2. You could bring more science to Talos Rowing - it is now done mostly via trial and error
   
3. You are making some valuable feature requests/suggestions

So yes - you could help a lot. For a while I have been wanting to have a data base of sensor data (Annotated Talos Rowing recorded sessions) at various stroke rates, boat configurations etc. I also need beta testers to try new Talos Rowing versions/features before releasing to the general public.

1-6 below need more chewing and discussion. As for 7:

7) It seems that my Samsung Galaxy SII could not reach decent GPS accuracy, Talos always shows the red indicator in his gps data. This is true always when other app are able to provide 99% gps accuracy (less than 5m error). Is it a bug?

there are 3 level indicators for speed/distance display based on the reported GPS sensor error range:

• < 2m = good (green)
  
• < 4m = so so (yellow)
• >= 6m = bad (red)

My reasoning was that for a slow-moving object such as a boat, with us being very fussy about a few seconds difference in our 500m time, an error range equal to more then half a stroke is bad. It shouldn't be red though for <5m. I will check.

 

--
You received this message because you are subscribed to the Google Groups "talos-rowing" group.
To view this discussion on the web visit https://groups.google.com/d/msg/talos-rowing/-/1T4dppch80sJ.
To post to this group, send email to talos-...@googlegroups.com.
To unsubscribe from this group, send email to talos-rowing...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/talos-rowing?hl=en.

[Tommi]

Very well!!
Recording and studying rowing data from high level rowers (high level= somebody who compete at Italian absolute level or higher) is a goal that require time, but I'm quite confident!  If you like the idea I can provide you something LIKE a scientific article to publish on Talos Rowing web site. I already written a paper about the analysis of longitudinal acceleration recorded using Talos Rowing, in two or three months, maybe, I can write something ad hoc to create a little technical wiki. This with the goal to explain only aspects that has a scientific proof. I'll told you the parameters that could be useful for future studies so that Talos can grow up and become a good instrument for coaches! I'll be pleased to test beta versions. If possible I'll try to record data with various boat configuration and stroke rates. Annotated rowing session is a little bit harder to do: Italian "rowing vocabulary and slang" is quite difficult and obviously different from the English one. A bad translation or a misunderstanding can make problems... I'll do my best =)

Can you make a step by step (maybe a simple video created by screen captures) tutorial about Talos Video Sync? I'll record some video and this will be a way to reduce misunderstandings in annotated session analysis.

Another idea: can you create  graph in which speed is function of stroke rate? This graph must be the result of a test: the rower must do 500m at 22 stroke per minute ---> 500 at 24 ---> 500 at 26 ---> 500 at 28/30/32 etc... Talos should guide the rower in doing that, maybe a special function in which the app records some 500 meters splits at a well definite stroke rate (displayed on the screen). This is really useful in analyzing rowers efficiency at various stroke rate. 

Can you check the csv export function of off line viewer? I'm doing something wrong? This is the result of exportation: PKɵyAgps.csv½™ÝŽÜ6…ïû,Æ@â?­g›.ŠI¤éEß¾‡3#ÏìZEöÂj‚l ÃóɦÈ

Thanks

[Tal Shalif]

On 26 November 2012 06:50, Tommi <tommaso...@gmail.com> wrote:

Very well!!
Recording and studying rowing data from high level rowers (high level= somebody who compete at Italian absolute level or higher) is a goal that require time, but I'm quite confident! 

Looking very much forward to such elite data. 'Golden' sample data would be the most valuable for possible future in-app rowing-hints and rower data analysis. However, any rowing data annotated by someone knowledgeable such as yourself would be useful for testing the basic Talos Rowing functionality.

If you like the idea I can provide you something LIKE a scientific article to publish on Talos Rowing web site. I already written a paper about the analysis of longitudinal acceleration recorded using Talos Rowing, in two or three months, maybe, I can write something ad hoc to create a little technical wiki.

There are these options:

1. Write your article on a blog or a dedicated site and link to it from the Talos Rowing site
2. Put it on the current Talos Rowing dev site and write such article/s Wiki style - it has a Wiki engine

Account holders on http://nargila.org/trac/robostroke should have write permission to the Wiki space.

This with the goal to explain only aspects that has a scientific proof. I'll told you the parameters that could be useful for future studies so that Talos can grow up and become a good instrument for coaches! I'll be pleased to test beta versions. If possible I'll try to record data with various boat configuration and stroke rates.

We need an army of testers - and as you noted, the more boat configurations the better.

Annotated rowing session is a little bit harder to do: Italian "rowing vocabulary and slang" is quite difficult and obviously different from the English one. A bad translation or a misunderstanding can make problems... I'll do my best =)

I am thinking about creating a standard form for session data submission with basic fields such as 

• boat configuration
• rower level
• wind/water conditions
• other comments

So - with the exception of 'other comments' - not much English language knowledge would be required.
 

Can you make a step by step (maybe a simple video created by screen captures) tutorial about Talos Video Sync? I'll record some video and this will be a way to reduce misunderstandings in annotated session analysis.

OK, will do

 

Another idea: can you create  graph in which speed is function of stroke rate? This graph must be the result of a test: the rower must do 500m at 22 stroke per minute ---> 500 at 24 ---> 500 at 26 ---> 500 at 28/30/32 etc... Talos should guide the rower in doing that, maybe a special function in which the app records some 500 meters splits at a well definite stroke rate (displayed on the screen). This is really useful in analyzing rowers efficiency at various stroke rate. 

Plotting the exported CVS stroke-rate data series and speed data series from such rowing data should be possible.

Can you check the csv export function of off line viewer? I'm doing something wrong? This is the result of exportation: PKɵyAgps.csv½™ÝŽÜ6…ïû,Æ@â?­g›.ŠI¤éEß¾‡3#ÏìZEöÂj‚l ÃóɦÈ

Thanks

The attachment of the gps.csv file did not show correctly. But it should look like this:

================= way.csv =====================

timestamp,distance,speed,accuracy

1352548807850,11.699664115905762,555000.0,5.0

1352548837866,40.35107421875,371000.0,5.0

.....

================= gps.csv =====================

timestamp,lat,long,alt,speed,bearing,accuracy

1352548800918,51.859315666370094,5.042589660733938,44.9000244140625,0.6880499720573425,61.0,5.0

1352548800919,51.859315666370094,5.042589660733938,44.9000244140625,0.6880499720573425,61.0,5.0

...

================= stoke_rate.csv =====================

timestamp,stroke_rate

1352548800902,63

1352548803753,21

...

These should be enough to create the above graphs - as long as your session recordings start and end exactly on the above mentioned 500m boundaries.