Auto Offside Detection in Football

Detecting offside automatically using adidas smartball and highlighting offside player using adafruit neopixel leds

Comment

Inspiration

The way football is played today is the same as it was a hundred years ago. With the rise of the world of smart technology like smart sensors and adidas smartball, we believe it is necessary to bring a change to the way the game is played, especially in important decisions in a match like offside which can cost a team a goal or even the match. We understand there are FIFA regulations in bringing technology to the game but even if it is not permitted to be used in the real game then it can definitely be used for the training of the players so that they can practice offside traps and hence can be a marketable product.

What it does

There are sensors attached to every player in the team (in their jerseys) giving their relative positions. Also we can get the position of the smart-ball relative to the players. Whenever a player from the attacking team(striker) is detected to have crossed the last defender of the defending team, then we start listening for kick events in the smart-ball. If at this point of time, a player kicks a ball and if this player is from the attacking team(we can identify the player who currently has the ball by detecting the player closest to the ball) then it is a potential offside. But in order to confirm the offside, the striker should also be receiving the ball. Thus finally we check whether the smart-ball is coming in proximity to the striker or not. If yes, then we immediately glow neopixels in the jersey of the striker player who committed the offside. All this happens in real time.

How we built it

We used proximity sensors, Genuino101, adidas smart-ball SDK and adafruit neopixels. We detected the kick event in the smart-ball using bluetooth LE and the smartball android SDK transfers this event to the Genuino101 using Bluetooth LE as well. Upon detecting offside we glow the neopixels.

Challenges we ran into

  • Lack of sufficient resources. The available proximity sensors were only the ultrasonic sensors. We needed more accurate sensors.
  • Only 2 member team.
  • No bluetooth beacons were available to use bluetooth as proximity sensors.
  • The smart-ball for the hackathon has very basic sensors.
  • For demo purpose, we are only simulating one defender and striker.

Accomplishments that we're proud of

  • Connecting the android smart-ball SDK to the sensor board was a bit of a hack but finally we managed to transmit the kick event to the board.
  • We were only 2 members but we managed to create a working prototype for our idea.

What we learned

  • Smart-ball android SDK.
  • Bluetooth Low Energy.
  • Arduino/Genuino101 sketch programming.

What's next for Auto Offside Detection in Football

  • By integrating better sensors into the smart-ball, this can become a marketable product which can be used in real matches.
  • It can also be used in training sessions to better train players for offside situations.
Share this project:

Updates