Around The NFL: Off-Season News and Rumors

[Callidus]

16 minutes ago, reo said:

 

You don't need a constellation. You just need need to track the x-y-z positions of the ball from a known point and probably it's orientation. 

 

The only other things you'd need might be the LOS and 1st down line but those could probably be set and saved by just storing the ball's position at specific times.

I was just using the terminology of the quoted post. Idk the technical  name for it but we aren't talking satellites. Either way the more receivers you have around the target the more accurate you should be able to get to a point. Obviously having 200k of them isn't going to let you see it to the micron.

[titanruss]

32 minutes ago, pat said:

 I’m curious how Zebra deals with the 15 channel limitation of UWB to track 2-3 tags per each 22 players.  There’s time sharing algorithms but it’s tricky.

 

It’s an interesting space for sure.

3-4 systems?

[reo]

1 hour ago, pat said:

You don't need a constellation. You just need need to track the x-y-z positions of the ball from a known point and probably it's orientation.
 

yeah, like a “constellation” of antennas in known positions perhaps?

 

there’s nice info here https://www.rfid-wiot-search.com/zebra-nfl-on-field-player-tracking

 

You don't need a constellation of antennas. Use an accelerometer. Zero it. Calculate the change in position in x-y-z. That's it. 

 

Not saying that's how they do it of course. I'll check the link later. Just boarded a plane to Miami. 

Edited May 23, 2024 by reo

[reo]

1 hour ago, Callidus said:

I was just using the terminology of the quoted post. Idk the technical  name for it but we aren't talking satellites. Either way the more receivers you have around the target the more accurate you should be able to get to a point. Obviously having 200k of them isn't going to let you see it to the micron.

 

Yeah that's different from what I'm talking about. That's more triangulation of positions. 

 

I'll catch up in a bit. Terrible signal at the gate and I'll have to go on airplane mode soon.

 

Fun conversation though. 

[Supernope]

 

[pat]

35 minutes ago, reo said:

 

You don't need a constellation of antennas. Use an accelerometer. Zero it. Calculate the change in position in x-y-z. That's it. 

 

Not saying that's how they do it of course. I'll check the link later. Just boarded a plane to Miami. 

That's dead reckoning.  It's notoriously susceptible to integration drift.  The missiles you work on likely have the most reliable inertial guidance systems available, but that's not what one gets in small package consumer grade electronics.

[Callidus]

57 minutes ago, pat said:

That's dead reckoning.  It's notoriously susceptible to integration drift.  The missiles you work on likely have the most reliable inertial guidance systems available, but that's not what one gets in small package consumer grade electronics.

They aren't putting a MIRV level tech into the footballs? Cheap ass nfl... I'd also like to known how much the football would weigh lol

[reo]

2 hours ago, pat said:

That's dead reckoning.  It's notoriously susceptible to integration drift.  The missiles you work on likely have the most reliable inertial guidance systems available, but that's not what one gets in small package consumer grade electronics.

 

I'm just talking about using a basic accelerometer. I don't really deal with guidance systems.

[pat]

1 hour ago, reo said:

 

I'm just talking about using a basic accelerometer. I don't really deal with guidance systems.

Yes, basic accelerometers such as those in phones drift quickly when doing dead reckoning.  The math is easy but sufficient precision in the sensor to be accurate is hard.  There was an interesting paper about using floor plans to correct for drift.  One could do something like that for football but at that point you are creating an algorithm to guess rather than actually measuring.

 

Pardon me for sharing my past work trauma, as far as offseason topics I figure it’s better than others.

[reo]

Just now, pat said:

Yes, basic accelerometers such as those in phones drift quickly when doing dead reckoning.  The math is easy but sufficient precision in the sensor to be accurate is hard.  There was an interesting paper about using floor plans to correct for drift.  One could do something like that for football but at that point you are creating an algorithm to guess rather than actually measuring.

 

Pardon me for sharing my past work trauma, as far as offseason topics I figure it’s better than others.

 

Interesting. I'll have to look into this more later.

[pat]

2 minutes ago, reo said:

 

Interesting. I'll have to look into this more later.

I think this was the paper I referenced 

 

https://www.scirp.org/journal/paperinformation?paperid=36346

[big2033]

 

[Callidus]

23 minutes ago, big2033 said:

 

My take away from this is the nfl for the most part was much higher on latham than the outside world

 This is the 5th i think, war room that confirms they expected latham to go 5 or 7

Edited May 24, 2024 by Callidus

[OilerTitanHybrid]

 

[AussieTitanFan08]