Skip to content
Tech Xplore — Robotics·· 18 days agoSignalEditorial score85

Soccer-playing humanoid robots score a major robotics goal for coordination

Soccer-playing humanoid robots score a major robotics goal for coordination

Summary

Researchers at Tsinghua University have developed a new control system for humanoid robots that enables coordinated soccer-playing behavior using onboard vision. The system was tested on a real robot and achieved an 80-90% success rate in scoring goals from the front half of the field, with a 64% reduction in response time compared to a rule-based system. The study was published in Science Robotics and demonstrates a significant advancement in embodied AI and robot coordination.

Editorial context

The study presents a novel control system for humanoid robots that enables coordinated soccer-playing behavior using onboard vision. The system was tested on a real robot and achieved an 80-90% success rate in scoring goals from the front half of the field, with a 64% reduction in response time compared to a rule-based system. The research was published in Science Robotics and demonstrates a step向

Source: Tech Xplore — Robotics · Read original article ↗

Loading article text…

What the source reports

Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.

Reported numbers

  • goals scored in competitions

    76

    View original evidence
    me for kickoff Then came the ultimate test—well, two to be precise—competing in international competitions. In 2025, the researchers' controller was used by the Tsinghua Hephaestus team in the RoboCup Adult-Size Humanoid League. So how did they do? They won the championship. At the 2025 World Humanoid Robot Games, the team used the same controller. They won that too! Across the two competitions, they scored 76 goals
    Open source S8
  • number of trials per point

    10

    View original evidence
    valuated the kicking success rate across 14 points on the soccer field, with 10 consecutive trials conducted at each point. Credit: Yushi Wang Written for you by our author Paul Arnold , edited by Sadie Harley , and fact-checked and reviewed by Robert Egan —this article is the result of careful human work. We rely on readers like you to keep independent science journalism alive. If this reporting matters to you, plea
    Open source S10
  • number of competitions

    2

    View original evidence
    me for kickoff Then came the ultimate test—well, two to be precise—competing in international competitions. In 2025, the researchers' controller was used by the Tsinghua Hephaestus team in the RoboCup Adult-Size Humanoid League. So how did they do? They won the championship. At the 2025 World Humanoid Robot Games, the team used the same controller. They won that too! Across the two competitions, they scored 76 goals
    Open source S8
  • paper: not_reported
  • code: not_reported
  • weights: not_reported
  • dataset: not_reported
Source excerpts and review record

Automatically extracted; no manual editorial approval recorded.

essfully kicked the ball into the goal around 80 to 90% of the time from positions in the front half of the field. The study authors also tested the controller across surfaces including grass, asphalt, rubber and soil. At the same time, it cut the time needed to reach and kick the ball by up to 64% compared to a state-of-the-art rule-based soccer system. Full explanation of the research process. Credit: Yushi Wang Ti

Open source S7

me for kickoff Then came the ultimate test—well, two to be precise—competing in international competitions. In 2025, the researchers' controller was used by the Tsinghua Hephaestus team in the RoboCup Adult-Size Humanoid League. So how did they do? They won the championship. At the 2025 World Humanoid Robot Games, the team used the same controller. They won that too! Across the two competitions, they scored 76 goals

Open source S8

valuated the kicking success rate across 14 points on the soccer field, with 10 consecutive trials conducted at each point. Credit: Yushi Wang Written for you by our author Paul Arnold , edited by Sadie Harley , and fact-checked and reviewed by Robert Egan —this article is the result of careful human work. We rely on readers like you to keep independent science journalism alive. If this reporting matters to you, plea

Open source S10

Source:Tech Xplore — Robotics · techxplore.com

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).