Researchers at the City University of Hong Kong have developed a soft sensor for sensitive measurement of shear forces applied to its surface. When attached to robotic grippers, the tactile sensing allows for fine control of grasped objects, enabling the robot to perform impressive feats, such as threading a needle and gripping an egg without breaking it. The researchers hope that the sensor may be useful for next-generation prostheses that can accomplish tasks that require very fine control. Human skin is adept at detecting shear force, which is the force that occurs when two objects slide over each other. When you feel something slipping out of your hand, your natural reaction is to tighten your grip, but robots (or powered prostheses) cannot typically do the same. This means that it can be difficult to perform certain tasks using robotic hands or grippers, and they tend to be a little clumsy. These issues have inspired this latest flexible sensor, which mimics the multilayered nature of the skin. The top layer is magnetized, and when a shear force is applied to it and deforms the sensor, it registers the resulting change in the magnetic field. It can also distinguish between shear force and the external force applied perpendicularly to a gripped object. “It is important to decouple the external force because each force component has its own influence on the object,” explained Yan Youcan, a researcher involved in the study. “And it is necessary to know the accurate value of each force component to analyze or control the stationary or moving state of the object,” The system uses machine learning to accurately characterize the position of the stimuli acting on it. “We have developed an efficient tactile super-resolution algorithm using deep learning and achieved a 60-fold improvement of the localization accuracy for contact position, which is the best among super-resolution methods reported so far,” said Dr Shen Yajing, another researcher involved in the study. “To the best of our knowledge, this is the first tactile sensor that achieved self-decoupling and super-resolution abilities simultaneously,” So far, the sensor has enabled a robotic gripper to complete some intricate tasks, suggesting that it may give prosthetic devices extra dexterity. For instance, the sensors allowed the gripper to hold an intact egg in place without breaking it, while a researcher attempted to pull it away. “The super-resolution of our sensor helps the robotic hand to adjust the contact position when it grasps an object. And the robotic arm can adjust force magnitude based on the force decoupling ability of the tactile sensor,” said Yajing. “This proposed sensor could be beneficial to various applications in the robotics field, such as adaptive grasping, dextrous manipulation, texture recognition, smart prosthetics and human-robot interaction,” Yajing added. “The advancement of soft artificial tactile sensors with skin-comparable characteristics can make domestic robots become part of our daily life.” Source
