As robots are increasingly asked to take on new tasks, researchers across the industry are constantly trying to improve their grip and dexterity. Many recently developed “soft robots”, however, have a flaw of their own: their added flexibility creates issues with accuracy, not to mention the ability to pick up the object. But MIT’s Computer Science and Artificial Intelligence Laboratory team might have successfully solved this issue.
The team’s soft robot hand, made from silicone rubber, has “bend sensors” that can analyze data based on their curvature. This allows the robot to compare new objects to ones it has previously picked up based on the similar data points. The robot is therefore able to pick up a variety of objects such as cups, eggs, CDs and tennis balls – all without damaging them. The CSAIL team hopes to add more sensors in the future so the robotic hand can distinguish an even wider variety of objects.
SO, WHAT DO YOU THINK?
Is this the next big step in soft robotics? Could this tech one day be used in an industrial setting?
Email us or leave your comments below.