Page 1 of 1

Tracking application for Canoe Polo

Posted: Mon Oct 10, 2016 2:25 am
by Sourcerer

7 years ago, I started with a project to develop a sports tracking solution.
The project is about CanoePolo:

One potential solution for the tracking I identified would be GNSS or LNS
Logger devices that could record coordinates with a horizontal accuracy of
less than 20 centimeters for 10 hours. (+ additional accelleration,
magnetic, ... sensors, but those are easy)
So I have been researching into RTK, but I could need something in
the price range of 100 USD per device.

The other idea is to track it with an aerial video from above:

Back then my market survey came to the conclusion that I would have to
wait a few years until the market could fulfill my requirements.

The NT1065 seems to be a good step in the right direction, and I guess
perhaps a GPS Logger based on NT1065 might become a viable solution for my

I would prefer to be able to buy an existing solution, but I also don't mind joining a project with hardware and software development.

Size constraints are not a big issue, we have about 20x20x10 centimeters
available in the boats, I would say. I would like to put 2 Loggers into a
boat, one in the front and one in the back. If that is too costly, I would start with one Logger in the front.
The sky depends on the place, it is mostly open I would say, but there are
even playgrounds in canals in a city in Italy, for example.
It is also played indoors, but that's seldom and there we can use video
tracking much much easier, so that's outside the scope.
It can work offline or online. We could provide Wifi, Internet for NTRIP, ...
We have enough time to calculate results offline. Usually the tournaments
are on a day, and we could post-process the results in our datacenter
during the night and have them available on the next day. That would be
fine for now.
Having everything in realtime would be nice, but it's not a requirement at the moment.
One playground is about 30x20 meters, for tournaments there could be perhaps up
to 5 playgrounds more or less next to each other. In Vienna we have 2
playgrounds for international tournaments.

Re: Tracking application for Canoe Polo

Posted: Mon Oct 10, 2016 6:16 am
by itsar

Right place to ask!

It will be difficult to make simple RTK because of large changes of roll of boats.
The phase center of antenna significantly degrades near antenna horizont.
You can track it by hyroscope but conventional receiver with raw output can
not correct measurements in real time.

I can imagine snapshot Logger with RTK. Snapshot Logger is capable of dumping
raw RF signal samples and download it later for processing. There are two examples:

You can make evaluation of such hardware device with our NUT4NT board and RA1 RTK-grade
Antenna. I think it is possible to make device in your size and price constrains in a large
production batch.

But the samples stream needs to be decimated about to 10 times to use practical size flash card.
So main question is does it possible for RTK-algorithm to converge with such object dymamics
and such losses by decimation.

May be someone on this forum already has a solution or you will find people who will make
system development.

Best regards,

Re: Tracking application for Canoe Polo

Posted: Mon Oct 10, 2016 7:09 pm
by Sourcerer
Yes, I guess that the professional antennas which are designed to be mounted upright all the time do not help much, my guess is that I would have to use simpler, more omnidirectional antennas, and that I have to live with more reflections and less signal and more noise.

Storage-wise my guess is that SD-cards will continue to grow in size for a given price in the next few years, so we could start with a smaller solution that can only capture 1 hour now, and replace the SD cards with larger cards a few years later. One match consists of 2 halftimes of 10 minutes each, which takes 30 minutes timeslots during a tournament. Exchanging the cards every hour should be doable.