BACKGROUND: MIT researchers are working on a very early version of intelligent, robotic helpers: a humanoid called Domo, who can grasp objects and place them on shelves or counters. Domo is the 'next generation' of two earlier robots built at MIT: Kismet, designed to interact with humans, and Cog, which could learn to manipulate unknown objects. Domo incorporates elements of both. A robot like Domo could help elderly or wheelchair-bound people with simple household tasks like putting away dishes. Other potential applications include agriculture, space travel, and assisting workers on an assembly line.
ABOUT ROBOTICS: Robots are made of roughly the same components as human beings: a body structure with moveable joints; a muscle system outfitted with motors and actuators to move that body structure; a sensory system to collect information from the surrounding environment; a power source to activate the body; and a computer "brain" system to process sensory information and tell the muscles what to do. Robots are manmade machines intended to replicate human and animal behavior. Roboticists can combine these basic elements with other technological innovations to create some very complex robotic systems. There are plenty of robots doing manual work on factory assembly lines, but while those machines can manipulate objects, they do the same thing, along the same path, every time. They aren't programmed to learn to adapt to new situations; Domo can. The philosophy behind the robot's development is that humans and robots can work together to accomplish tasks that neither could do alone.
HOW DOMO WORKS: Domo can 'see' everything that happens in front of it thanks to large blue eyes equipped with cameras that scan the room. The cameras feed visual information to 12 computers that analyze the input and decide what to focus on. This is important because for a robot to function in a real-world human environment, such as a kitchen, it must be able to ignore clutter and focus only on certain stimuli. Domoýs visual system is attuned to unexpected motion. For instance, locating human faces is critical for social interaction and people are often in motion. When Domo spots motion that looks like a face, it locks its gaze onto it. Once Domo's gaze is captured, the human can issue verbal commands, to find a shelf, for example. The robot will scan the room for a shelf and then reach out a hand to touch the object to make sure it is really there. If an object is then placed in its hand -- such as a bag of coffee beans -- the robot will reach up and place the object on the shelf.
WIGGLE ME THIS: To get a feel for the size and weight of any object placed in its hand, Domo wiggles it a little. The movement is minor, yet critical to the robot's ability to accurately place it on the shelf. Domo is programmed to learn about the size of an object by focusing on its tip, such as the cap of a water bottle. When the robot wiggles the tip back and forth, it can figure out how big the bottle is and decide how to transfer it from hand to hand, or to place it on a shelf. Domo can also sense when a human is touching it, thanks to springs in its arms, hands and neck that can sense force and response to it. If too much force is applied, the robot will voice its displeasure by saying "ouch!"
The Institute of Electrical and Electronics Engineers, Inc., contributed to the information contained in the TV portion of this report.