MIT researchers are developing robots that could work side by side with humans in factory production lines.
The researchers have developed an algorithm that allows robots to learn and adapt to humans, understand how individuals work and anticipate their needs as they work.
The robots will basically help a mechanic carry out a worker’s tasks effortlessly, by picking up the worker’s preferences in a synchronised manner without affecting the worker working.
The development is the work of Julie Shah, the Boeing Career Development Assistant Professor of Aeronautics and Astronautics at MIT. Shah leads the Interactive Robotics Group in MIT's Computer Science and Artificial Intelligence Laboratory. The aeronautics researcher is more interested with the use of such robotic assistants in airplane manufacturing plants.
According to Shah, having a robot assistant perform small tasks such as picking tools and materials for a mechanic will greatly reduce time wasted moving from one point to the other to pick the items. The robot’s algorithms enable it to learn people’s way of working and predict their next move or what they require. But people have dissimilar working mannerisms. Shah hopes the robots will someday undergo training with humans, understand their working and make effective assistants.
When finally on the factory floor, the robot will easily distinguish mechanics and engage the appropriate work mannerisms for each mechanic. The robots will use radio-frequency identification tags as is standard currently to distinguish mechanics.