Robots help out—the way we want them to!
Both humans and robots work in manufacturing plants however, they don’t usually work alongside one another. Robots are most often used in repetitive, exhausting or hazardous work, while people are needed for tasks that require finer skill and detail. An earlier post on this blog discussed some of the ways robots have started working alongside humans but there are of course some important issues to consider if we really want to make this kind of collaboration work.
Humans are individuals and we like to do things our own way. According to Julie Shah, an Assistant Professor of Aeronautics and Astronautics at MIT, robots need to display an almost seamless understanding of how they can help people. To give robots the ability to do this, Shah and her team at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a new algorithm that allows a robot to quickly learn how people prefer to do particular tasks so that it can adapt accordingly to give them a hand.Shah stresses that without this ability, people are likely to get frustrated and forego any assistance from the robot which could potentially increase their efficiency on the task.
Watch the video above to see how the algorithm works in a test case involving spar assembly, a specific part of airplane manufacturing. The algorithm is also being used in simulations to train robots and humans to work together; findings will be presented at the Robotics: Science and Systems Conference in Sydney, Australia this July.