Puck passing the goal line detection

[Seraph 240]

I have always had this idea too. I can tell you it's not even debatable that the technology is available. It's just a matter of how much you want to spend to make it as accurate as possible. Right now people are performing surgeries with robots over the internet and controlling rovers in real time on Mars. This type of a project is relatively simple and probably attainable as a finals project in a college engineering class. Doesn't matter if the puck goes in sideways or not--you could also teach it to only count it once the whole thing is over the line.

I think it would be a good idea to implement but probably won't make it high on the list of priorities any time soon. I don't think the league feels there is too much of a problem with missed calls.

[Hack & Whack Rule! 160]

Not a bad idea. Here's the thing, though. Right now, the puck has to cross the line completely, showing white ice between the puck and the line. There is currently something that detects that. It's called a pair of f***in' eyeballs! Any measure implemented by the NHL is going to be f***ed into the stone age by incompetent half-wits who like to play God by re-writing very recent history. Seriously, how do you screw up the "white ice between the puck and the line" thing?

[BlueMonk 102]

Since the puck is 3 inches in diameter and the goal line is 2in wide, you would also know if it crossed the line by only signaling a goal when the center of the puck is 2.5in past the center of the goal line (where the receivers would be positioned in the middle of the goal posts).

You could use this method but I don't know how you adjust for the fact the puck isn't spherical. What if it's on end instead of flat, or some degree of rotation in between? The center of the puck would need to be varying distances from the virtual line in those instances. I'm not sure it's so simple to determine whether it's over the line or not.

Edit: After I replied, I see others have raised the same point.

[VM1138 1,921]

The only problem I could see, if I'm understanding you correctly, is that the puck has to completely be over the goal line. I can see how it would detect when it reached it, but not if it was completely over it.

I'm not very bright, though. So you may be on to something.

The War Room would have nothing to fear. They still have intent to blow.

Well, they could paint the strip an inch or so BEHIND the goal line, or far enough back that the entire puck would have to be past the goal line for it to reach the strip and thus trigger the light.

[kook_10 1,705]

You could use this method but I don't know how you adjust for the fact the puck isn't spherical. What if it's on end instead of flat, or some degree of rotation in between? The center of the puck would need to be varying distances from the virtual line in those instances. I'm not sure it's so simple to determine whether it's over the line or not.

Edit: After I replied, I see others have raised the same point.

you could have a balanced pair of tags in the puck. they're plenty small enough. with a 3 point sensor system it would see everything.

[Textbook 1]

The official dimensions of an NHL hockey puck are 1" thick and 3" in diameter. Simple solution: Stick an RFID directly in the center of the puck. Rig every net with an RFID sensor at 0.5 inches past the back edge of the red line for vertical-oriented pucks - this RFID sensor would have to be at all 4 corners. Then, place a second set of sensors 1.5 inches past the back edge of the red line for horizontal-oriented pucks. How difficult is that?

Edit: the only problem would be if the sensors were placed on the net and the net moved at the same time there was a close play.

Edited January 17, 2010 by Textbook

[Hockeytown0001 7,652]

I doubt this will happen.

Because it might, you know, actually protect the integrity of the game instead of the officials' off-ice bets.

[dragonballgtz 273]

I would rather have them find a way to put a camera underneath the ice.

Edited January 17, 2010 by dragonballgtz

[EastLansingNative 1]

Someone could probably do something with RF tags, which are ridiculously small these days, but there would be a few big challenges. You are right, it would be a challenge to ensure the puck's dynamics aren't changed by putting something really tiny in the middle. Strength of signal through the rubber of the puck could be an issue, being frozen may be an issue, and certainly being rugged enough to take a 100+ mph slapshot would be an issue. If there were antennae in the goalposts and the crossbar then the position could be triangulated, so it would know if the puck went over. Since the puck is 3 inches in diameter and the goal line is 2in wide, you would also know if it crossed the line by only signaling a goal when the center of the puck is 2.5in past the center of the goal line (where the receivers would be positioned in the middle of the goal posts).

I work with barcode scanners, wireless networking, and RFID for a living. Been at it for over a decade.

On paper, yes one could design a system, but in the real world, it would not achieve the accuracy or consistency that the game requires.

http://en.wikipedia.org/wiki/Radio-frequency_identification

There are many types of tags; active, passive, battery assist passive. There are many different frequencies, and some dual band tags.

There is a TON of, what we call in the industry, "VooDoo" involved with this technology. It is still very much "bleeding edge". (Not very accurate or consistent.)

The absolute best you're going to get with this technology is about a 12" x 12" square, and the smaller you make it, the less chance you have of getting good reads when something actually enters the read zone.

RFID is ok for large round types of spaces (Think: forklift driving into the back of a semi truck trailer to load pallets of product) because you've got large directional antennas on the reader that are bombarding the general vicinity with RF to "exite" the tag, and initiate a response.

In the diagram below, if the gray rectangles are your antennas, and the blue shaded areas are your read zone, you're looking at my 30-second version of the best case scenario... ...not good.

(And for all you syntax snobs out there: In our industry, insects have antennae, Systems Engineers install antennas)

And for anyone that wants to suggest putting an antenna under the ice: Water absorbs RF, and ice even more so.

The other main issue is that RF goes where it's not supposed to sometimes, and every time the puck landed on the back of the net, or bounced off a post, or came within 3"-6" of the net area, you'd have your good goal indicator triggered.

As the technology, accuracy, and consistency of this technology improve, I'll let you know if I come across a feasible option.

As far as some sort of contact based stripe/paint whatever, that's out of my realm of expertise, and I cant speak to it.

[EastLansingNative 1]

I would rather have them find a way to put a camera underneath the ice.

I like this idea, and it would certainly be more feasible than an RF based system.

Plus, how cool would it be to get a shot of the Zamboni blades spinning directly above the lens.

[Guest Howard He Do It?!]

I work with barcode scanners, wireless networking, and RFID for a living. Been at it for over a decade.

On paper, yes one could design a system, but in the real world, it would not achieve the accuracy or consistency that the game requires.

http://en.wikipedia.org/wiki/Radio-frequency_identification

There are many types of tags; active, passive, battery assist passive. There are many different frequencies, and some dual band tags.

There is a TON of, what we call in the industry, "VooDoo" involved with this technology. It is still very much "bleeding edge". (Not very accurate or consistent.)

The absolute best you're going to get with this technology is about a 12" x 12" square, and the smaller you make it, the less chance you have of getting good reads when something actually enters the read zone.

RFID is ok for large round types of spaces (Think: forklift driving into the back of a semi truck trailer to load pallets of product) because you've got large directional antennas on the reader that are bombarding the general vicinity with RF to "exite" the tag, and initiate a response.

In the diagram below, if the gray rectangles are your antennas, and the blue shaded areas are your read zone, you're looking at my 30-second version of the best case scenario... ...not good.

(And for all you syntax snobs out there: In our industry, insects have antennae, Systems Engineers install antennas)

And for anyone that wants to suggest putting an antenna under the ice: Water absorbs RF, and ice even more so.

The other main issue is that RF goes where it's not supposed to sometimes, and every time the puck landed on the back of the net, or bounced off a post, or came within 3"-6" of the net area, you'd have your good goal indicator triggered.

As the technology, accuracy, and consistency of this technology improve, I'll let you know if I come across a feasible option.

As far as some sort of contact based stripe/paint whatever, that's out of my realm of expertise, and I cant speak to it.

Great post! It's very refreshing to hear from someone who knows quite a bit about the subject and can add some scientific reasoning to the debate.

[Wings_Dynasty 267]

The NHL can't use technology for good reasons. Tell them you want the puck to glow. They can do stupid s*** like that.

[RedWingsRox 614]

Don't waste your intellectual energy people, NHL is not reading LGW board for ideas to better the league.

[Guest Howard He Do It?!]

Don't waste your intellectual energy people, NHL is not reading LGW board for ideas to better the league.

While that is true, it is still an interesting subject to talk about.

[SweWings 45]

Don't waste your intellectual energy people, NHL is not reading LGW board for ideas to better the league.

No, but a journalist/blogger might. If enough noise is made they might be asked questions about it. They might not take those questions seriously - no surprise! - but they'll have to consider the subject at least.