Engineering news
Designed to improve robotic grip and handling of fragile, slippery or asymmetric objects, the system was developed by a team led by researchers at the University of Surrey, who hope that it could “pave the way for safer, more reliable automation across industries ranging from manufacturing to healthcare”.
Conventional robots are trained to hold objects by relying solely on grip force, which can be ineffective or even damaging to delicate items. Humans, on the other hand, have much more dextrous control of objects thanks to an ability to predict how they will move.
“If you imagine carrying a plate that starts to slip, most people don’t simply squeeze harder – they instinctively adjust their hand’s motion by slowing down, tilting or repositioning to stop it from falling,” said Dr Amir Ghalamzan, associate professor in robotics and lead author of a new study on the work.
“We’ve taught our robots to take a more human-like approach, sensing when something might slip and automatically adjusting their movements to keep objects secure. This could be a game changer for future automation, from handling surgical tools in healthcare and assembling delicate parts in manufacturing to sorting awkward packages in logistics or assisting people in their homes.”
The team quantified the effectiveness of ‘trajectory modulation’ for slip prevention in both humans and robots. By doing so, they showed that a predictive control system powered by a learned ‘tactile forward model’ allowed robots to anticipate when an object was likely to slip by continuously analysing its planned movements.
Robots are generally trained to apply the minimum level of force to objects to prevent damage, said Professor Venky Dubey from Bournemouth University, who was not involved in the work. “Many times we have to do some delicate manipulation while it is stationary or moving. So in that case, I think the slip sensation, or object slip, plays a very vital role,” he said to Professional Engineering.
“This is particularly very useful when some kind of teleoperation has to be done by a robotic system for robotic surgery, where we are not close to the object location… if we get a sensation of the object slipping then we can manipulate accordingly.”
Combining slip sensing with normal force sensing could “definitely” make an important contribution to better teleoperated manipulation, the robotic hand and sensor specialist added, “particularly if you are dealing with arteries, or some kind of mass inside human bodies, then I think that will add realistic sensation of what they're performing.” Predictive control systems could also be used in purely robotic surgical operations in future, he added, by integrating with force sensing.
The Surrey researchers also demonstrated that the system works on objects and movement paths that it was not trained on, highlighting its potential to generalise effectively to real-world settings.
“We believe that our approach has notable potential in a variety of industrial and service robotic applications, and our work opens up new opportunities to bring robots into our daily life. We hope our findings will inspire future research in this area and further advance the field of robotics,” said Dr Ghalamzan.
The work was carried out in collaboration with the University of Lincoln, Toshiba Europe’s Cambridge Research Laboratory, Arizona State University and the Korea Advanced Institute of Science and Technology.
The study was published in Nature Machine Intelligence.
Want the best engineering stories delivered straight to your inbox? The Professional Engineering newsletter gives you vital updates on the most cutting-edge engineering and exciting new job opportunities. To sign up, click here.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.