Teaching Commercial Robots How to Adapt


Many commercial robotic arms perform "pick and place" tasks - the arm picks up an object in one location from an assembly line and places it in another. Usually, the objects are positioned so that the arm can easily grasp them and the appendage that does the grasping may even be tailored to the objects' shape. General-purpose household robots, however, would have to be able to manipulate objects of any shape, left in any location. Today, commercially available robots don't have anything like the dexterity of the human hand. Students in the Learning and Intelligent Systems Group at MIT's Computer Science and Artificial Intelligence Laboratory have demonstrated how household robots could use a little lateral thinking to compensate for their physical shortcomings. MIT senior Annie Holladay describes how her algorithm helps the robot adapt by using both of its arms instead of just one.