For several decades, various types of artificial intelligence kept shocking the world. Robots could people in highly competitive games and then quickly destroyed their human competitors.

    AI long ago mastered chess, the Chinese board game Go and even the Rubik's cube, which it managed to solve in just 0. 38 second.

    Now machines have a new game that will allow them to humans: Jenga, the popular game in which players remove pieces from an increasingly unstable tower of 54 blocks, placing each one on top until the entire structure would .

    A newly released video from MIT shows a robot developed by the school's engineers playing the game with surprising accuracy. The machine is equipped with a soft gripper (夹子), a force-sensing wrist and an external camera, allowing the robot to detect the tower's the way a human might do

    Unlike in purely recognitive tasks or games such as chess or Go, playing the game of Jenga also requires mastery of physical acts such as pushing, pulling, placing, and arranging pieces. It must interactive physical operation, where you have to touch the tower to learn how and when to move blocks.

    Imitating it is rather difficult, so the robot has to learn in the real world, by working with the real Jenga tower. Recently, a relevant research was published in the journal Science Robotics. Researchers say the robot demonstrates that machines can learn how to perform certain tasks through actual touching instead of relying heavily on visual . That physical is significant, researchers say, because it provides further proof that robots can be used to perform tasks, such as separating recyclable objects from landfill trash and assembling consumer products.

    In a cellphone assembly line, the felling of any component is coming from force and touch rather than vision. To become an accomplished Jenga player, the robot did not require as much repetitive practice as you might imagine. Hoping to avoid reconstructing a Jenga tower thousands of times, researchers developed a method that allowed the robot to be trained on about 300 games. Researchers say the robot has already begun facing off against humans, who remain players—for now.