The robot was designed by researchers from the Korea University in Seoul and the Berlin Institute of Technology.

Dubbed "Curly", the robot was claimed to assess the target and allowed to throw one stone prior to matches to assess the conditions.

Curly is actually two robots, with one assessing the position of the stones while the other releases the stone.

Researchers said the robot won three of four matches against expert curling teams.

"The game of curling can be considered a good test bed for studying the interaction between artificial intelligence systems and the real world," researchers said in the journal Science Robotics.

"In curling, the environmental characteristics change at every moment, and every throw has an impact on the outcome of the match.

"Furthermore, there is no time for relearning during a curling match due to the timing rules of the game.

"Here, we report a curling robot that can achieve human-level performance in the game of curling using an adaptive deep reinforcement learning framework."

"Our proposed adaptation framework extends standard deep reinforcement learning using temporal features, which learn to compensate for the uncertainties and nonstationarities that are an unavoidable part of curling," researchers added.

"Our curling robot, Curly, was able to win three of four official matches against expert human teams [top-ranked women’s curling teams and Korea national wheelchair curling team (reserve team)].

"These results indicate that the gap between physics-based simulators and the real world can be narrowed."

The robot did not have sweepers during the matches.

The full research has been published in Science Robotics.