Science Buddies: "Ask an Expert"

I hope you all know that FIFA just released the new technology which will be used for soccer games. The point is to make sure when ball crosses the goal line which will be notified by this technology. In past, some teams were screwed because referees couldn't see everything 100%. So my project is to built a goal with some sort of sensor that will recognize whether the ball crossed the line or not. If it does, my speaker will say "GOAL". Another thing I am going to add is video camera which will record the motion of starting to the neding point of the ball, so that later, one can do reply and see the motion in slow motion. Also I am building soccer player in which I will build shotgun.
My question for you is if you have any good comments of how to approach this project. Thanks a lot. http://news.bbc.co.uk/sport1/hi/football/4148155.stm

Hi buffalobuki!

Not an answer, just a pointer with background to one of your questions:

http://news.bbc.co.uk/sport1/hi/football/4148155.stm

Think of how burglar alarms work. Any of those technologies can be adapted to this:

Infrared beams

Ultrasonic beams

Weight of the ball on the grass (only works for shots on the ground)

Disturbance of RF fields (neat technology, but probably not practical for this)

Some very accurate RF technology involving triangulating the position of the ball with a RF emitter in the ball. (Probably not practical).

For all of the beam technologies, you would need to account for the very real possibility of the goalie breaking the beam but the ball remaining out of the goal.

Just a few thoughts. I'm sure there are many other possibilities.

I know that the article mentions no video (as in humans watching instant replays), but video motion detection is another thought. The goal mouth should look the same frame by frame on video. But if the ball crosses the goal mouth, the video image will change. I had a similar system on my pool camera when my kids were young that would alarm if the video image on the pool camera changed by a certain number of pixels.

Hi buffalobuki!

Oops! I provided the same URL as you already had, BIG sigh. This is going to require more research; the first time through that was the best URL I found. I'll let you know if I come up with anything better.

Hopefully a fellow Science Buddy will be able to help, too.

The more I think about this topic, the more I think that video motion detection might be an elegant solution.

If you could make your soccer goalie in a single monochromatic color, your could exclude that color from the consideration of the video image motion (sort of how blue screens work in movies). Then, assuming a white soccer ball, any white motion over a certain number of pixels past the goal line would signify a goal.

Anyway, my $0.02. Please let us know what you choose and how it works out.

I would think the main problem would be telling the difference between the ball crossing the goal and something else crossing the goal, e.g., a hand or foot. Putting one of those anti-shoplifting tags on the ball might work; you know, the kind they put of DVD that set of the alarm at the front door if they have not been demagnetized.

johnsteele wrote:I would think the main problem would be telling the difference between the ball crossing the goal and something else crossing the goal, e.g., a hand or foot. Putting one of those anti-shoplifting tags on the ball might work; you know, the kind they put of DVD that set of the alarm at the front door if they have not been demagnetized.

The problem there, I would think, would be the location of the tag. Assuming that it is placed in the center of the ball, unless that center passed the goal line, it wouldn't register. But only a fraction of the ball needs to cross to count as a goal, so I don't know if that would work.

Of course, you could place the sensors 1 ball's radius outside the goal, so maybe that can work.

This really is an interesting problem.

Hi All!

Here are some websites discussing various aspects of soccer goal-line technology:

computer chip in soccer ball with sensors in field:
http://news.bbc.co.uk/sport1/hi/football/4206450.stm

four high-resolution cameras mounted high above the field:
http://sports.yahoo.com/sow/news?slug=a ... &type=lgns

instant replays, general discussion:
http://news.bbc.co.uk/sport1/hi/footbal ... 074506.stm

Hi all,

I feel that these high tech technologies that uses "sensors" (be it infrared disruption, sensors on the ground, etc) can falsely trigger if for example it were not the ball that went through the goal line and onto the grass behind it. A goal keeper's hand can falsely trigger these sensors if the ball almost hit the goal line (but didn't) but the goal keeper's fingers did go in because he was trying to deflect the ball.

Then the types that uses magnetic and RF field needs a device actually implanted in the ball itself. That would actually make the ball not "pure". That is the rules to what the ball is would need to be changed and it would then mean a revolution to traditional soccer rules.

So I think that a "video" monitoring with a human being as the final judge should be the best route to take (just like some of the experts have mentioned above).

Plus we wouldn't want a "non-human" to make decisions for us humans would? Especially when it comes to a goal in a soccer game that may only happen once during the whole duration of the game.

Just an opinion.

George:

I couldn't agree more about the human factor. But the initial article had said that FIFA refused to use video instant replays for the method, so they have chosen some other electromechanical method of doing this.

I would love to know what they chose.

The NBA just tried changing the basketballs to microfiber instead of leather. Big disaster, and they just retreated, tail between their legs, to the old leather balls. I wonder if soccer players are as picky as basketball players?

The video motion sensor technology can work without human intervention, and without the goalie's body crossing the goal line affecting this (except if the goalie covers up the ball totally so that it can't be seen by the cameras. I don't know what would work in that situation, outside of RF). For a science fair experiment, it is fairly simple, and would work most of the time. Probably more than enough for his purposes.

BTW, have we lost you buffalobuki with our discussion? Still out there listening?

I am working on project where I have to "connect" SONY controller with a FPGA Spartan 3E board through microController and then connect the same FRGA board to the ping pong gun that should be able to aim the ping pong ball up/down/left/right towards specific target. That ping pong gun will be build with 4 dc stepper motors; 1 for aiming left/right, 1 for aiming up/down, 1 for moving up/down, 1 for moving left/right. So I have to write Verilog code for that gun. Also that gun should be able to move in any direction, and to shot the ball. All that will be done by user who will be using that SONY controller. So my qusetion is whether you have any Verilog code that can realize that process ?

buffalobuki wrote:I am working on project where I have to "connect" SONY controller with a FPGA Spartan 3E board through microController and then connect the same FRGA board to the ping pong gun that should be able to aim the ping pong ball up/down/left/right towards specific target. That ping pong gun will be build with 4 dc stepper motors; 1 for aiming left/right, 1 for aiming up/down, 1 for moving up/down, 1 for moving left/right. So I have to write Verilog code for that gun. Also that gun should be able to move in any direction, and to shot the ball. All that will be done by user who will be using that SONY controller. So my qusetion is whether you have any Verilog code that can realize that process ?

Please only ask the question once... I think you should post your question in the Computer science forum, as I answered in the other thread.