Researchers at the Georgia Institute of Technology have created a system that makes a human-controlled robot more "intelligent," and improves the amount of control that a human user has over it. It incorporates a number of sensors that are placed on the user's arm to read muscle information, and help the robot to anticipate the user's intentions. The system has been developed to improve safety and efficiency in manufacturing plants.

The research cites the example of car manufacturing workers who have to hang car doors on hinges, using a lever to guide a robot that carries the door. Whilst the power-assisting device used sounds practical, it is not necessarily easy to use.

"It turns into a constant tug of war between the person and the robot," explains Billy Gallagher, a recent Georgia Tech PhD graduate in robotics who led the project. "Both react to each other's forces when working together. The problem is that a person's muscle stiffness is never constant, and a robot doesn't always know how to correctly react."

The Georgia Tech system eliminates any "confusion" in the robot by monitoring the operator's muscle movements and sending the data to a computer. The system judges the person's intention and adjusts its interaction accordingly, resulting in safer and easier use.

The main aims of the research are to improve safety, time and efficiency in manufacturing plants, and it is expected to benefit industries including automobile manufacture, aerospace and military. The system will continue to be developed using a US$1.2 million National Robotics Initiative grant supported by a National Science Foundation grant and will focus on better understanding the "mechanisms of neuromotor adaptation in human-robot physical interaction."

"Future robots must be able to understand people better," says Jun Ueda, Professor at the Woodruff School of Mechanical Engineering. "By making robots smarter, we can make them safer and more efficient."

Watch the video below to find out more about how the system works.

View gallery - 4 images