IoT. Google AI Soccer Table

By Matteo Trevisan

Three years ago, I proposed to get a foosball table for the office so that we can enjoy it while taking a break during the working day. This idea turned out to be much better than I had imagined. Not only did we have several high-energy, adrenaline-filled breaks, but it allowed us to team up in different groups and get to know each other better.

After a year of great, intensive matches, naturally, we asked ourselves, “which is the strongest team at Ennova Research?” We enthusiastically accepted this challenge and used our personal skills to study and apply new technologies to make our leisure time “smart”.

The Google AI Soccer Table project was developed by five teams, each with a specific research objective. To determine the best team of the tournament, we set up the following items :

• face recognition and image capturing of players
• goal detection through infrared (IR) sensors
• linking goals to the appropriate team
• display of acquired data on a screen
• data processing and determination of the winning team

Each team performed the following tasks:

After about a year of research and development on the project, we were thrilled with the results and presented the project at the Google DevFest event in Venice with an overwhelmingly enthusiastic response from both the players and the audience. Personally, I was very excited to be involved in this project and it increased my interest in understanding and applying new technologies in a dynamic and friendly context.

At the end, after a period of constant use and testing, the technology confirmed what I always knew… I am undoubtedly the strongest player at Ennova Research!