Trace GPS Test

[Horton]

@dlokshin came out to Bakersfield to do some testing last weekend. We need to make some tweaks before we get high quality slalom data but here is a tiny taste. You will see the speeds all look spot on.

 

Yes I know it is upside-down. When the bugs are worked out we will turn the camera over.

https://youtu.be/eEmLzkju68c

[Adam Caldwell]

@dlokshin - Nice work! Looks awesome!

 

Does this also log roll/pitch/yaw ?

 

In addition to Max and Average Speed, anther parameter of interest to me is minimum speed. Is that an option ?

 

Very cool, looking forward to using one sooner then later hopefully!

[MattP]

Sweet stuff @Horton and @dlokshin! I'm looking forward to see how the product progresses! It will be a great insight into what we skiers (think we) are feeling on the water.

[Horton]

@adamhcaldwell don't scare @dlokshin. I told him we need good speed data first.

[bishop8950]

Whats with the silly short slalom ski and why not go around the buoys??

 

Very cool. I have always worked for science companies where having good data is always interesting and critical to success. For my slalom skiing I have been happy to live without a lot of data. This said, really curious to see what we can learn. Will be interested in how this tracks slalom speed changes to start. Especially if tied to position in the course.

[AdamCord]

This would be HUGE for skiing from a ski design perspective, but more importantly as a training tool. We know that more speed is better, but to be able to track it precisely, compare to other skiers, compare after fin changes, etc. would be amazing. I'd use it every set.

 

What would be amazing is if we could even use it as a coaching tool. If there were a user interface that can actually tell you where you're slow/fast/late/early and the probable causes. With enough data I'm sure we'd be able to see patterns that make this clear.

[bbirlew]

@AdamCord,

Do you have access to an android device? I'll share the interface I've been working on to display this type of info. I'm trying to design it general enough to import data from the Trace, as well as the Xensr, and other relevant info (strain gauge, etc).

I would eventually like to incorporate video as well.

I have it set up as a 'ski log book' that will import your set data and store things like fin settings, notes, etc with it. You can then compare the data with previous sets.

Drop me a line if it sounds like somthing you would like to play around with.

 

[AdamCord]

@bbirlew I unfortunately do not have an Android device. @Horton is nice enough to let me know how stupid I am for it on a biweekly basis.

 

If there's any other way I can use it, I'd love to!

[Horton]

@bbirlew are you aware that the Trace API is now public?

[bbirlew]

@horton,

@dlokshin informed me a while back that changes were on the way, but last time I checked out their website I couldn't find any new info.

 

I unfortunately don't have a trace to test with, but I do have a xensr to play around with. Now I'm waiting on some API info from them and some warmer water! Though I do fully intend to support Trace data as well.

 

I'll definitely plot any new data you send me though and use it for testing and development.

[dlokshin]

@bbirlew Lots of people using the API now (including a NASA contractor!). You can find more info here: http://developers.Traceup.com

[dlokshin]

NP This is a pull API, but we have a push API as well, which you can email me about it you need it.

[Horton]

@dlokshin @Rico @Colebrah @Wolfeie and I did some more testing yesterday. Thanks to @Wolfeie for use of his boat. We should have some video to show in a couple of days.

 

I saw some things in the speed data yesterday I really didn't expect.

 

We mounted a unit on a helmet and we also mounted a unit on a ski . The speed of the ski at apex is radically different than the speed of your head at apex. Now that I think about it - it is obvious that your upper body is moving much slower than your lower body as the ski comes back under the rope.

 

The other thing that really confused me until I talked to @AdamCord was that the data showed a skier going just slightly faster five or ten feet after edge change than at center line. This is the result of the skier staying connected and the ZO catching up. @Colebrah was picking up about one MPH from center line to the whitewash. It would be fun to change the ZO settings and see if you could see the Delta.

 

More testing asap. Who around here can write some C# or Ruby or C++

 

 

[bbirlew]

C# or c++.

Or tell me what you want to see and I'll work it into my android app.

 

Drop me a line if you want to play around with it.

[Horton]

@bbirlew since I do not know what is possible so it is hard to exactly know what to ask for.

 

The first step to a technical tool is a simple dashboard that gives the max speed for gate pullout, pull out and then each wake crossing for the pass. It is not as fun as the video overlay but if you told that with one ski my average speed at centerline was always faster than another ski, that would be amazing.

 

I think min speed data might also be interesting but that is more more dependent on where the Trace is mounted. I would think that the higher the min speed is on a gate turn in the better.

 

@AdamCord I assume highest pull out speed and highest min speed at turn in is all the goal of the GUT gate?

 

 

[Horton]

Here is some overhead data from Saturday. I honestly do not think exact GPS position is nearly as interesting for us as speed data but this is pretty darn cool.

 

http://goo.gl/erZZq0

[Horton]

This is so freaking exciting. Check out this video. It is not perfect but so so so close. The time/data sync on the video is off by about 20 feet but it is SPEED DATA and I believe it is darn good data. The data to video sync correction is not super hard. With luck the folks at Trace will get us a corrected video later this week.

 

As I said somewhere above - the data / video sync is fun but I am actually more interested in peak speed for each crossing and perhaps min speed at gate turn in. The video to speed thing is icing for how I want to use this tool.

 

I really think this could change the way we all think about slalom.

 

https://youtu.be/fo9j1HEVlsY

[A_B]

I saw the trace on FB. This is cool stuff!

So can you denote a color difference change for direction? Would be great to see the difference between the same line going down and back. Is it due to onside offside? Is it from being ahead of the "middle" on way to the next buoy?

 

Can this be applied to boat path and is it accurate and timely enough to judge boat path?

 

Cool!!!

[SkiJay]

Fascinating progress @dlokshin, congrats! At what stage are you with plotting and manipulating graphic traces of speed and location for each pass? Being able to overlay traces and manually step through them point by point is the tool I'd find most helpful. (I'm on Android)

[Than_Bogan]

Neat! I hope this manages to go somewhere.

 

About the "surprise":

 

Every model I've ever tried to create, and also @gator1's overhead view data, all expect the peak speed to occur after the centerline. It has nothing (well, ok, very little) to do with ZO and everything to do with vectors. Before the centerline, your tangential speed (i.e. perpendicular to the rope) is partially in the opposite direction from the boat. After the centerline, it is partially in the same direction as the boat.

 

So even if the magnitude of your tangential speed is dropping as you cross the centerline, the sum of the vectors (tangential + boat) will slightly increase in magnitude as your tangential's component in the boat direction changes from negative to positive.

[Chef23]

Very cool information. It is interesting to see the speeds into the 50s. It will be cool when we can see the actual speeds overlayed in the appropriate place in the course.

[DefectiveDave]

I'm definitely excited! Even with the sync issue it's still informative. I'm also glad to see Trace stepping up to support our sport. After asking my wife, I hope to be ordering one this evening, haha.

[rab]

Cool stuff, @Horton what was the camera mount setup for the slalom video?

[Horton]

@rab the camera mount is in @Colebrah's mouth. Like a scuba mouth piece. We call it the swallow cam.

 

Future video on this project will be mostly Wakeye from boat

 

[klindy]

A bit off topic but this also looks like an interesting device. A waterski version would be way cool.

KickStarter

 

[Horton]

@klindy watch your long links please

[klindy]

Sorry @Horton I was on an iPhone and couldn't figure out how to shorten it.

[Horton]

@klindy you are far from the only offender. Try https://goo.gl/ it works well from your phone if you are willing to go through a few extra steps.

[disland]

This is awesome. Video is cool but I believe a graphical view of speed vs time would be most useful. You could then easily convert that to an acceleration graph. This would provide at a glance how much load was being managed from pass to pass.