Gophers football: GPS systems help keep players on track

Donate To GopherHole
supadupafly

supadupafly

Active member
Joined
Apr 13, 2011
Messages
3,181
Reaction score
1
Points
36
Per Marcus Fuller (St. Paul Pioneer Press):

A year ago, Gophers football players were so excited about being on the cutting edge of athletic performance tracking that they couldn't keep it to themselves.

Coach Jerry Kill didn't quite understand how the technology would help yet -- and wanted to hold off making a big deal.

But part of the University of Minnesota's success during an 8-5 season was due to their ability to measure the speed, exertion and fatigue of players in practices and games.

http://www.twincities.com/popular/ci_26282306?source=most_viewed
 

Very cool stuff. Nice quote Ced: "Everybody would like to have nice facilities but coach does things that will make us better," Thompson said.
 


Pretty cool. I didn't know Cedric Thompson was that fast.
 

Nitpicking, but I thought GPS wasn't actually that accurate. Even now that they've "unlocked" the precision, I thought it still had a +/- of like 1 or 2 meters. Wouldn't that matter when calculating the difference between someone running 22 and 21.7 MPH?

I thought RFID or something else more "local" would be more accurate.

Sincerely,
I am not an expert.
 


BarnBoy said:
Nitpicking, but I thought GPS wasn't actually that accurate. Even now that they've "unlocked" the precision, I thought it still had a +/- of like 1 or 2 meters. Wouldn't that matter when calculating the difference between someone running 22 and 21.7 MPH?

I thought RFID or something else more "local" would be more accurate.

Sincerely,
I am not an expert.
Click to expand...

The GPS that most people use isn't very accurate, but the technology can be incredibly accurate with the right equipment.

I'll let wiki do the explaining:
http://en.wikipedia.org/wiki/Differential_GPS
 

Barnboy, thinking the same thing. To get sub-meter accuracies there is a lot of processing of the data which is only as good as the programmer. This is similar to the hand vs laser timed 40 times. I believe these sensors operate at 10 hz as well.

The technology is probably good for showing trends from a baseline, eg after an injury or from beginning of a game but I'm not sure of the utility of one-time measurements. Multiple trials could fine tune the accuracy of 40 times but I doubt they would match the accuracy of laser. The accelerometer data could be interesting.

Michael Vick supposedly runs a 21.63 mph and Usain Bolt 27 mph.

It would be interesting to see the claims of the the manufacturers.
 


As I understand it, the accuracy improves with sampling rate. i-phones sample at 1 Hz (one time per second) and are generally accurate to 5 meters. The GPS devices that the football teams use are sampling at 5-10 Hz, and are more accurate. The trade off is the amount of data you collect is 10 times higher, so storage becomes more of an issue.
 



FSN: Gophers use technology to monitor players' workload

per FSN:

"You've got to be careful with what you do with the kids," Kill said. "Some of them are banged up a little bit, but that's how camp goes. You're looking at a lot of young kids. We've got a system right now and kind of understand what we're doing with how hard we work them and so forth."

Kill mentioned the team's use of the Catapult system, which, according to its website (catapultsports.com), is "the most used secret in sports." The system collects data during the Gophers' practice -- things such as how many miles players have run, how fast they move, etc. -- and allows coaches and trainers to monitor the progress.

The company's site says that a dozen NFL teams and 18 Division I universities -- including Minnesota -- use the Australia-based Catapult system. The University of Nebraska is the only other Big Ten school in that group.

The data gathered by Catapult can be vital for teams like the Gophers, especially during the grind of camp. While Minnesota has remained relatively healthy through the first two weeks of practice, monitoring the athletes' workouts helps keep them fresh.

"We get to Thursday, we're two weeks out," Kill said. "We've got to scale back and concentrate and focus on some things, condense everything, and get people sharp and better. But there were some good things happening out there today."

http://www.foxsports.com/north/stor...technology-to-monitor-players-workload-081214

Go Gophers!!
 


I am enjoying all of the GPS nerdery. Hope it is alright to throw my expert knowledge in the ring.

You guys are focused on using position to estimate velocity and are probably missing that velocity is most accurately estimated with GPS using the Doppler of the carrier signal directly. Since the L1 carrier wavelength is only a little over 19 cm and the satellite-to-receiver timing accuracy is on the order of nanoseconds or better post-acquisition, the 3D velocity you can estimate is actually very accurate, especially with differential GPS corrections. Because of this, GPS receivers sold commercially have the so-called Wassner restrictions in place so they cannot be used to track at high speeds and be used in, say, guidance of a missile. Cheap GPS receivers don't support use of the Doppler, so they estimate velocity/speed using the traditional divide-change-in-position-by-change-in-time method, which is dominated by position errors that are likely to be quite large given the large buildings nearby that Differential GPS likely can't remove.

Since human beings, and especially practicing football players are highly dynamic, the comment about the faster sampling is a very astute one. You probably want 5 Hz or better sampling.

Differential GPS is almost certainly used here, although it usually requires post processing of the data (which to me the article suggests). Otherwise you have other large hurdles to overcome - most notably data link.

Apologies for the nerdgasm.
 

AeroGopher said:
I am enjoying all of the GPS nerdery. Hope it is alright to throw my expert knowledge in the ring.

You guys are focused on using position to estimate velocity and are probably missing that velocity is most accurately estimated with GPS using the Doppler of the carrier signal directly. Since the L1 carrier wavelength is only a little over 19 cm and the satellite-to-receiver timing accuracy is on the order of nanoseconds or better post-acquisition, the 3D velocity you can estimate is actually very accurate. Because of this, GPS receivers sold commercially have the so-called Wassner restrictions in place so they cannot be used to track at high speeds and be used in, say, guidance of a missile. Cheap GPS receivers don't support use of the Doppler, so they estimate velocity/speed using the traditional divide-change-in-position-by-change-in-time method, which is dominated by position errors that are likely to be quite large given the large buildings nearby.

Since human beings, and especially practicing football players are highly dynamic, the comment about the faster sampling is a very astute one. You probably want 5 Hz or better sampling.

Apologies for the nerdgasm.
Click to expand...

Well duh...tell us something we don't know.

You probably should have posted this in the www.gopherhole.com/boards/exlain_it_to_me_like_im_five forum!

Damn! My fake link is too long to show the whole thing!

Cool stuff. I know nothing about it, and I'm glad we are one of the few B10 teams that utilize it, and I hope it stays that way.
 



You must log in or register to reply here.



Top Bottom