This technique, little explored to date, is suitable for handling fragile or irregularly shaped objects, as well as working near humans without risk.
Professors at the Universidad Carlos III de Madrid (UC3M) and the Universidad Pública de Navarra (UPNA) have developed a soft robotic gripper capable not only of picking up and holding objects, but also of manipulating and repositioning what it holds between its fingers without losing its grip. As the researchers explain, this is a capability that almost no soft gripper possesses today, offering multiple applications ranging from handling fragile or irregularly shaped objects to operating safely near people.
The authors of the study, recently published in the scientific journal Bioinspiration & Biomimetics, are Carlos Relaño Gibert, professor in the Department of Engineering at the UPNA, and Concepción A. Monje Micharet, Alberto Rodríguez-Sanz, and Lisbeth Mena, professors at the UC3M Department of Systems Engineering and Automation's RoboticsLab. Both the assembly and the operation of the gripper can be viewed in videos available on the YouTube platform.
Advantages over rigid industrial robots
As the authors of the study explain, traditional industrial robots are made of rigid components and are highly precise, but also very clumsy: they need to know the exact shape of the object they are going to pick up beforehand, and if they make a mistake, they can break it or injure someone in their path. “Soft robotics proposes the opposite: building robots with flexible materials that automatically adapt to whatever they touch, just as our fingers conform to a glass without us having to think about it. This makes them particularly suitable for manipulating fragile or irregularly shaped objects, and for working safely near people,” they detail.
The novelty of the newly developed gripper lies in the fact that, until now, most grippers “know how to pick up and hold, but not much more.” “This gripper features three soft fingers, each with three degrees of freedom, which in addition to bending can rotate on their own base. By coordinating these movements, the gripper rolls the object between its fingers and reorients it without ever letting go,” notes the research team.
“The analogy would be what we do when we pick up a jar and turn it in our hand to read the label without putting it down on the table, or the motion of unscrewing a lightbulb. An action that comes naturally to us without thinking, but which can be enormously difficult for a robot. And that is precisely where the difference lies: most soft grippers know how to grab an object, but they do not know what to do with it afterward. Ours can reposition it without releasing it. It is not about imitating a human hand, but about giving robots a bit more of that dexterity we use constantly to manipulate objects,” they illustrate. Furthermore, the gripper design is modular. That is, the three fingers can be assembled and disassembled separately, so if one breaks, it can be replaced without having to change the entire gripper.
Results show that the system is capable of maintaining a highly reliable grip. Highly diverse objects were used in testing, such as a water bottle, a cube, a pack of tissues, a fidget spinner, a 3D-printed figurine, a rubber duck, several stuffed toys, a screwdriver, and an artificial rose.
Regarding its applications, the gripper can be used for any task where a robot needs to adjust how it holds an object without having to release it and pick it up again: from orienting a part prior to assembly to handling fragile or irregular items in sectors such as food processing, logistics, or laboratory settings. It can also be particularly relevant in collaborative and assistive robotics, where robots must continuously adapt to their environment and work safely alongside humans.
The study is funded by the ADAPTA and SIROCO projects, from the Ministry of Science, Innovation and Universities through the Spanish State Research Agency (AEI), in addition to FEDER funds from the European Union. Specifically: The SIROCO Project, PID2023-147343OB-I00, is funded by MICIU / AEI / 10.13039/501100011033 and FEDER, EU. The ADAPTA Project, PLEC2023-010218, is funded by MICIU / AEI / 10.13039/501100011033.
Video: https://youtube.com/shorts/3dZYaSnARjs