Does anyone else recall that childhood friend that used to cheat in rock, paper, scissors? Let’s say that the next turn in Super Mario 64 was on the line and you decided to settle a score with a quick best of three series. But of course, that little jerk hesitated when you threw rock and quickly changed his scissors into paper.
Well, the Ishikawa Lab in Tokyo has developed a “Janken” robot that mimics this on the millisecond level. And it wins every single time.
Here’s how it works:
In this research we develop a janken (rock-paper-scissors) robot with 100% winning rate as one example of human-machine cooperation systems. Human being plays one of rock, paper and scissors at the timing of one, two, three. According to the timing, the robot hand plays one of three kinds so as to beat the human being.
Recognition of human hand can be performed at 1ms with a high-speed vision, and the position and the shape of the human hand are recognized. The wrist joint angle of the robot hand is controlled based on the position of the human hand. The vision recognizes one of rock, paper and scissors based on the shape of the human hand. After that, the robot hand plays one of rock, paper and scissors so as to beat the human being in 1ms.
So, it’s a big fat cheater.
Check it out in action below:
On a side note, let me indulge my robot paranoia for a second. As a society, we’ve already gone too far and developed robots that can do push-ups and even sweat, ones that can morph in order to fit into tiny spaces, and ones that can learn based on historical actions.
Now they are going to be able to recognize our attack before we barely move a muscle? We’re screwed.
[Via Ubergizmo]