Ri-one’s Fusion-fueled rise to “world’s best” in RoboCup

“Ri-one,” a certified student organization at Ritsumeikan University, continues to challenge the international stage at RoboCup—a prestigious global robotics competition—by leveraging cutting-edge robotics and AI technologies.

RoboCup isn’t your typical robotics competition. Its most famous challenge is the RoboCup Soccer competition, where teams of autonomous robots play soccer matches against each other. At RoboCup 2024 in the Netherlands, Ri-one claimed their first-ever championship title in the Small Size League (SSL) Division B, competing against nine teams from eight countries.

This remarkable achievement was underpinned by Autodesk Fusion, which played a central role in all stages of development of their soccer-playing robot, from drive mechanism design and circuit board layout to structural optimization and simulation. The team created eight identical robots with more than 4,000 small, precise parts and had zero breakdowns during the competition.

Leading the SSL team is Riku Miyazaki, a third-year student in the Department of Electrical and Electronic Engineering at the College of Science and Engineering. Established as an independent student organization in 2016, Ri-one currently comprises around 50 students—mainly from the College of Information Science and Engineering and the College of Science and Engineering—of which about 30 belong to the SSL team.

“In Division B, AI analyzes video footage from overhead cameras installed around the field and remotely controls six robots to play soccer matches,” he says. “As technology has advanced, it's no longer enough to just roll the ball—we now see matches being won or lost based on whether the robots can kick it upward and pass it. Our victory this year was largely due to the perfection of our kicking mechanism.”

Daiki Tomioka, a student in the Department of Robotics, has been deeply involved in robot fabrication since his junior high school days.

“I started participating in RoboCup Junior in middle school, and I joined Ri-one because a senior member who had competed with me in RoboCup Junior was already there,” he says. “In fact, many of us on the current team, including those seniors, used to compete against each other as rivals in different teams.

After entering university, they reunited and formed a true “dream team,” working together to solve several design challenges.

“To implement the kicking mechanism, we needed to secure internal space, which pushed us to miniaturize the undercarriage,” Tomioko says. “We developed an integrated assembly combining a brushless motor driver, internal gear mechanism, and omni-wheel, significantly reducing the overall size that allows us to fit in the kicker system.”

Fusion’s parametric modeling played a key role in optimizing the omni-wheels.

“By inputting parameters, we created a model where we could freely adjust dimensions like the wheel diameter and sub-wheel width to find the optimal design,” Tomioko explains. “The fact that students can achieve such advanced designs is one of Fusion’s greatest strengths.”

Despite winning the world championship in the SSL Division B, Ri-one’s journey is far from over. Their next goal is to enter Division A, which presents a significantly higher level of difficulty. In this division, even greater levels of robot autonomy and team coordination are required. Very few university student teams compete here; most participants are researchers or professional engineers.

“In Division A, each robot is equipped with its own camera and must independently perceive its environment, make decisions, and act using AI. On top of that, the entire team needs to act cohesively,” Miyazaki says. “The number of robots increases from six to a maximum of eleven, demanding more complex data processing and control.”

To tackle these new challenges, Ri-one’s design team has started using Fusion’s generative design capabilities. Traditionally, design processes have relied on the designer’s experience and creativity. But with generative design, Fusion can automatically generate thousands of design alternatives by setting conditions such as load-bearing requirements, material selection, installation space, and manufacturing method.

“If we're going to mount cameras on the robots, minimizing weight and making efficient use of internal space will become more important than ever,” Tomioka says.

“With generative design, we can achieve complex, efficient geometries that would be difficult for humans to conceive,” he adds. “We can fully utilize the available space while maintaining structural integrity. This machine learning-powered design method truly represents the future of manufacturing.”

Fusion’s generative design supports a wide range of manufacturing methods, from 3D printing to CNC machining. Even for student projects, it enables realistic, production-ready design, seamlessly connecting ideation to prototyping and manufacturing.

Armed with real-world experience on the global stage, they are ready to take on even tougher challenges. And standing by them throughout this journey is Fusion—empowering their imagination, enabling their technical growth, and helping them shape what’s next. Fusion will continue to be a trusted partner on their path to the next “world’s best.”

Get free educational access to software, learning content, and other resources.