IEEE Robotics and Automation Letters Best Paper Award: Study by Sant’Anna School of Advanced Studies Receives the 2025 Honorable Mention

IEEE Robotics and Automation Letters Best Paper Award: Study by Sant'Anna School of Advanced Studies Receives the 2025 Honorable Mention

A prestigious recognition has been awarded to research conducted at the Sant'Anna School of Advanced Studies in Pisa. A study carried out within the BRAIR Lab (BRAin-Inspired Robotics Laboratory) of the Institute of BioRobotics was selected from more than 1,700 papers and received the 2025 Honorable Mention at the IEEE Robotics and Automation Letters Best Paper Award.

The award ceremony took place in Vienna during the IEEE ICRA 2026 Conference. Representing the Sant'Anna School was Enrico Donato, co-author of the study together with Egidio Falotico, Scientific Coordinator of the BRAIR Lab.

The Algorithm That Enables Robots to Recognize Objects Through Touch

The study presents an innovative AI algorithm that allows a robot to recognize objects solely through touch, without the aid of vision. Researchers demonstrated that a soft robotic gripper equipped with sensors can identify more than sixteen different objects by analyzing both the pressure exerted during grasping and the way the fingers deform when coming into contact with the object.

At the core of the research is an algorithm based on recurrent neural networks, designed to interpret over time the tactile signals collected by the gripper. Rather than relying on a single contact event, the system learns from the evolution of the grasp and combines multiple sources of sensory information, significantly improving its ability to distinguish between objects of different shapes and sizes.

One of the most interesting findings is that combining multiple sensory modalities is far more effective than relying on a single sensor. Furthermore, the robot can progressively increase its confidence in object classification by repeatedly grasping the same object from different postures, similarly to how a person manipulates an unfamiliar object to better understand its shape.

This work represents an important step toward more autonomous robots capable of operating and interacting in complex environments where vision may be limited or insufficient. The research also paves the way for future technologies in which robots will be able to explore and understand the world through a sense of touch that increasingly resembles that of humans.