A small team of roboticists at Robotic Systems Lab, ETH Zurich, in Switzerland, has designed, built and tested a four-legged robot capable of playing badminton with human players.
In their study, published in the journal Science Robotics, the group used a reinforcement learning-based controller to give the robot the ability to track, predict and respond to the movement of a shuttlecock in play, demonstrating the feasibility of using multi-legged robots in dynamic sports scenarios.
Badminton is a sport similar to tennis, the main difference being the use of a shuttlecock rather than a tennis ball. The goal is the same: to hit the shuttlecock over a net placed midcourt to an awaiting opponent.
Playing badminton requires skillful footwork to maneuver quickly into a desired position and dexterity of the arms and hands to accurately hit the shuttlecock and send it over the net to a desired location. Giving a robot such abilities required some adaptations.
The main adaptation was giving it four legs instead of the two used by humans. The configuration gives the robot much more stability and flexibility in movements.
To give their robot such abilities, the researchers gave it a stereo camera and a dynamic arm. They also gave it a reinforcement-learning-based controller to allow the robot to position itself and to respond appropriately to the shuttlecock.
Helping the robot was the implementation of what the team describes as a “perception noise model” to compare real-world data from the camera with data from the information database it built during training. This allowed the robot to engage in human-like badminton player activities, like following through on strikes and pitching (tilting its base backward or forward) to keep the shuttlecock in view.
-
Credit: 2025 Yuntao Ma, Robotic Systems Lab, ETH Zurich
-
Badminton robot. Credit: 2025 Yuntao Ma, Robotic Systems Lab, ETH Zurich
Testing showed the training paid off; the robot, which the team has named ANYmal-D, was able to move around the court reasonably well and maintain rallies with human players for up to 10 shots.
More information:
Yuntao Ma et al, Learning coordinated badminton skills for legged manipulators, Science Robotics (2025). DOI: 10.1126/scirobotics.adu3922
© 2025 Science X Network
Citation:
Four-legged robot plays badminton with humans (2025, May 29)
retrieved 29 May 2025
from https://techxplore.com/news/2025-05-legged-robot-plays-badminton-humans.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
