Learning From Ghosts - How AI And Machine Learning Are Changing Sports

Learning From Ghosts - How AI And Machine Learning Are Changing Sports

It was an odd but effective analogy that the Manchester United players heard that day from their manager. “I remember going to see Andrea Bocelli, the opera singer. I had never been to a classical concert in my life. But I am watching this and thinking about the coordination and the teamwork, one starts and one stops, just fantastic. So I spoke to my players about the orchestra - how they are a perfect team.”

Sir Alex Ferguson, who won 38 trophies during his 26 years in charge at Old Trafford, recalled that particular pre-game talk to Anita Elberse, a professor at Harvard Business School, as part of a case study she created about his demanding but successful management style, albeit of a sports team rather than a company.

The symphony metaphor is appropriate for most team-based, invasion-type sports as only the unified efforts of all players creates the desired result, whether it be harmonious music or consistent victories.  "To me, teamwork is the beauty of our sport, where you have five acting as one,” said Mike Krzyzewski, all-time wins leader in college basketball, who sounds much like Phil Jackson, owner of 11 NBA Championship rings, "the strength of the team is each individual member. The strength of each member is the team."

During a game, one player’s movement influences not only his teammates’ proactive adjustments but also the reaction of his opponents. A ball carrier’s cut to the left instead of the right changes the dynamics of both teams. At the end of the game, it’s interesting to know each individual’s analytics, like distance covered, passes completed and shooting percentage, but it is vital to visualize the coordinated movement of the team to truly understand how games are won and lost. The outside defender, small forward or right winger may have had a particularly good or bad day, but their net effect on the ensemble is what matters.

For decades, coaches have relied on game film to recall and explain what happened. Watching the action on video gives a richer, realistic recap of the motion that static statistics can’t provide. More recently, combining film with a numerical analysis offered two important but distinct assessments that still requires coaches to integrate. Today’s attempts to bring together the analog fidelity of film with the digital accuracy of analytics has stalled. Annotating video clips with play data, which allows for easier searches and context specific stats, helps but provides no way to apply advanced tools, like artificial intelligence (AI) and machine learning, to the thousands of micro movements and positional changes of players throughout a game.

Read More

The Next Madden Game Frontier?

(Graphic courtesy of Oregon State University)
For all of you Madden 12 junkies out there, I've got a new post over at Axon Potential on some current artificial intelligence research being done at Oregon State University.  They are attempting to teach a computer system to watch an OSU football game and be able to identify, categorize and then suggest plays in a football simulation.

Certainly a tall order, even for some humans, but they've had some initial success with a small playbook of twenty passing plays.

According to the lead researcher, “This is one of the first attempts to put several systems together and let a computer see something in the visual world, study it and then learn how to control it,” said Alan Fern, an associate professor of computer science at OSU. “Football actually makes a pretty good test bed, because it’s much more complicated than you might think both visually and strategically, but also takes place in a structured setting. This makes it quite analogous to other potential applications.”

It seems the developers at EA Sports may have a head start on play selection AI, based on my poor record against the Madden gods.

Thanks for making the jump to Axon Potential to read the rest of the story.

Artificial Intelligence Gets A Kick From Soccer Androids

The world's best players may soon be facing a new challenge from football playing robots, which their creators claim will be able to play and beat a human team. According to new research in WIREs Cognitive Science, building robots to play football is driving the development of artificial intelligence and robotic technology which can be used for roles including search and rescue and home help.

The author, Claude Sammut, from the ARC Centre of Excellence for Autonomous Systems in Sydney, reviewed the technology demonstrated at the RoboCup international robot soccer competition which this year took place in Singapore. Competitions have become a popular way for motivating innovations in robotics and provide teams of scientists with a way of comparing and testing new methods of programming artificial intelligence (AI).

"Football is a useful task for scientists developing robotic artificial intelligence because it requires the robot to perceive its environment, to use its sensors to build a model of that environment and then use that data to reason and take appropriate actions," said Sammut. "On a football pitch that environment is rapidly changing and unpredictable requiring a robot to swiftly perceive, reason, act and interact accordingly."

As with human players football also demands communication and cooperation between robotic players and crucially requires the ability to learn, as teams adjust their tactics to better take on their opponents.
Aside from football the competition also includes leagues for urban search and rescue and robotic home helpers which take place in areas simulating collapsed buildings and residential homes, revealing the multiple use of this technology.

While a football pitch layout is structured and known in advance, a search and rescue environment is highly unstructured and so the competition's rescue arena presents developers with a new set of challenges. On the football pitch the robots are able to localize and orientate themselves by recognising landmarks such as the goal post, yet in a rescue situation such localization is extremely difficult, meaning that the robot has to simultaneously map its environment while reacting and interacting to the surroundings.

In the home help competitions the robot is programmed to recognise appliances and landmarks which will be common in most homes, but in addition to orientating themselves they must react and interact with humans.

As the robotic technology continues to develop the rules of the competitions are altered and made harder to encourage innovation, it is the organisers' aim that this will drive the technology to a level where the football playing robots could challenge a human team.

"In 1968 John McCarthy and Donald Michie made a bet with chess champion David Levy that within 10 years a computer program could beat him," concluded Sammut. "It took a bit longer but eventually such programs came into being. It is in that same spirit of a great challenge that RoboCup aims, by the year 2050, to develop a team of fully autonomous robots that can win against the human world soccer champion team."

So while, for the moment, football players can focus on beating each other to lift silverware, tomorrow they may be facing a very different challenge.

Source: Claude Sammut. Robot soccer. Wiley Interdisciplinary Reviews: Cognitive Science, 2010; DOI: 10.1002/wcs.86 and  Wiley - Blackwell


See also: Soccer Robots Are Getting Smarter At RoboCup and Soccer Robots Getting Smarter At RoboCup