
RoboCup and the UPennalizers
RoboCup
RoboCup is an international robotic soccer competition whose aim is research and education. The dream is:
“By mid-21st century, a team of fully autonomous humanoid robot soccer players shall win against the winner of the most recent World Cup.”
RoboCup brings in participants from all over the world. The primary aim of the competition is to encourage the growth of robotics development across the globe and to provide a test-bed for new and innovative ideas.
Team History
The Upennalizers were one of the first participants in the original Sony Aibo league (renamed the Standard Platform 4-legged league in 2004), and participated in all the international competitions between 1999-2006. Our team finished among at least the quarter-finalists in every one of those eight years. After a short hiatus in 2007-2008, we reformulated the team with a group of undergraduate and graduate students and competed in the new Standard Platform League (SPL) and the Simulation League. In 2010, we will be competing in the SPL, Simulation League, and the Humanoid League.

Highlights
• Second place finish at US Open, 2009
• Quarter-final finish at Osaka, 2005
• Second place finish at US Open, 2005
• Fourth place at Lisbon, 2004
• Second place at Padua, 2003
Team background
Our team consists of students from diverse fields of study. At RoboCup 2009, our team was composed of students pursuing computer science, robotics, mechanical engineering, math and finance. This year, we have recruited more graduate and undergraduate students in order to strengthen our team. Since we will also be collaborating with Virginia Tech University to build our own robots that will compete in the Humanoid League, our team now also consists of several students with a background in embedded systems.
Challenges
The aim of RoboCup is to address some of the challenges in robotics in a competitive setting. Some of these include humanoid locomotion, robust computer vision, and localization in noisy environments. To get a feel for the difficulties involved, consider a typical task that a soccer playing robot must achieve. It must first survey the scene it is looking at and detect the ball. It must then be able to deduce its own position and orientation solely from the information that it receives from its camera. Then, the robot has to walk to the ball, a task that requires it to continuously balance itself using its inertial measurement unit. Finally, it has to balance itself on one leg and use the other leg to kick the ball. And all this must be done in only a few seconds!
UPennalizers, University of Pennsylvania
