Georgia Tech's Using Human Arms To Make Robots More Accurate

Where robotics innovation is concerned, the manufacturing sector is a major driving force, particularly where heavy machinery is concerned. The reason for this is simple - such machinery has the capacity to be extremely dangerous to humans, and with the increasing focus on automation, all it takes is a simple miscalibration to kill or disable a worker. For this reason, researchers, inventors, and innovators all over the world are constantly coming up with new ideas, concepts, and inventions to make manufacturing smarter, better, and most of all, safer.

The latest development to that end comes from the Georgia Institute of Technology, which is devising a control system which, upon completion, will make robots both safer and considerably more intelligent. In order to help this process along, researchers at the Institute are using arm sensors on human beings in order to train robots to mimic, respond to, and even predict the movements of their squishier peers. Ultimately, they hope this will allow man and machine to work together seamlessly. 

This is no easy task, of course. Humans and robots move in fundamentally different fashions. For example, when a human wants to stop a certain movement and hold a lever still, their arm muscles stiffen and contract, pulling back. This in turn confuses the robot, which has no way of working out whether that contraction is a signal to change direction or simply the force of muscle contraction. The robot reacts, which causes the human to react further, entering into what essentially amounts to a feedback loop where neither party is quite accomplishing what they originally set out to do. 

"It turns into a constant tug of war between the person and the robot," explained Billy Gallagher, a recent Ph.D. graduate in robotics and project head in a school news release. "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 solution, continued Gallagher, is to give the robots more information instead of simply expecting it to react to its human counterparts on the fly. That's where the arm sensors come in. By allowing the robot to read its operators muscle movements, it can actively adjust its own position to adapt. 

The whole project is being carried out at the behest of the National Robotics Initiative, which has gifted the Georgia Institute of Technology with a $1.2 million grant. Ultimately, the project hopes to create smarter robots, able to interact more smoothly with their human counterparts. What this means is that for individuals in the manufacturing industry, their jobs are slated to get a whole lot safer - and a whole lot easier, too. 

The manufacturing sector is one of the primary driving forces of innovation in the robotics industry - but it isn't the only area in which the technology can be used. The drive to make the workplace both more efficient and safer by building smarter robots has given us innovations that can make robotics as a whole better, and our robot partners more capable of adapting to and interacting with their human creators.