Have you ever imagined, like human beings, robots will be able to recognize itself in mirror? Thanks to advanced robotic technology. Some computer scientists at Yale University, US have developed a humanoid robot called Nico that can recognise itself in mirror.
Nico is an upper-body humanoid robot. It is developed at Yale University’s Social Robotics Lab, under the direction of Brian Scassellati, an associate professor in Department of Computer Science at Yale University.
Nico is the first humanoid robot which is capable of identifying its own reflection. It has a simple head along with two cameras per eye for wide and narrow fields of vision, a pair of arms, and a torso. Microphones for speech recognition are attached on its body. Its body has a total of 22 degrees of freedom (head x 6, 2 arms x 6, right hand x 2, back and waist x 2). Nico uses sixteen computers for image processing and other tasks.
Nico is programmed to compare what it sees in the mirror with its own movements. If the mirror shows an arm moving and Nico’s own arm is moving at the same time, Nico can conclude that it is looking at itself. It is pretty much interesting. First, Nico’s sensor data detects the arm’s movements and then compare this arm’s movement with its camera images. The vision side of the equation can estimate the arm’s position to within two centimeters in any direction. If the sensor and image data matched, Nico assumes that it is looking at a reflection. If Nico doesn’t see any arm movement, then Nico presumes that it is looking at another robot.
Justin Hart, a PhD student who is leading the research said “It is a spatial reasoning task for the robot to understand that its arm is on it not on the other side of the mirror.”
For such programming, researchers had to to do a lot of hard work. They programmed the robot to identify its own body parts, to distinguish between animate and inanimate objects, to track others’ gazes, to recognize postural shifts and read body language, and more.
At present, the robot is in its experimental stage called the ‘mirror test. So far the robot has been programmed to recognise a reflection of its arm. But Mr. Hart wants it to pass the “full mirror test”. Mr Hart along with his supervisor Brian Scassellati, will publish a complete result in the spring.