Robot Vision Gets Better

Seeing something and knowing what it is and how you can use it can take a human being less than a fraction of a second to compute. However when it comes to a robot, the whole process is overseen by a complex algorithm that takes in input from a camera and then categorizes what is shown.

Robot vision is not so much true vision as a program that gets an image and tells the robot what the object in the image is. Then it also lets the robot know the uses that the object that has been sighted. This is a skill that all robots that hope to work in households in the future must master. At the Massachusetts Institute of Technology researchers are working on making this easier for the robots to do.

Lawson Wong, a graduate student in electrical engineering and computer science has been working on a new algorithm that can make robot vision for future household robots better. After several attempts and improvements the researcher has come up a new algorithm that shows a lot of promise. The objects that a robot is commonly likely to encounter are sorted into specific groups to aid faster recognition of the object. The tests of the science project have been promising.