A new electronic skin that gives robots a sense of touch

In everyday life, people constantly rely on their sense of touch to interact with the world. Yet, for robots, perceiving delicate forces has long been a stumbling block. Traditional tactile sensors often come with complex designs and require time‐consuming calibration, which limits their practical use in real-world applications.

To address these challenges, the researchers turned to nature for inspiration. Drawing on the remarkable abilities of human skin and the self-decoupling property of the Halbach array, they developed, after two years of research, a soft magnetic sensor that can decouple forces in three dimensions. This work was supported by the “Défi Clé Robotique centrée sur l’humain" (Human-Centered Robotics Key Challenge) funded by Occitanie Region, France and the CNRS.

The sensor’s design is deceptively simple: it is made up of three layers. On top is a flexible magnetic film that deforms when touched, causing a change in the magnetic field. An intermediate elastomer layer acts as a cushion, while a bottom layer—equipped with a printed circuit board and Hall sensors—detects the variations in the magnetic field. This configuration enables the sensor to measure both normal (perpendicular) and shear (parallel) forces independently.

Their recent paper, published in Nature Machine Intelligence, details several exciting demonstrations of this patented technology. For example, by integrating the sensor into an artificial knee joint, the researchers were able to monitor how forces are distributed during movement—a result that could help clinicians better understand and treat joint issues. In another experiment, the sensor was used to guide a robot in making coffee. By applying gentle pressure and sliding motions on the sensor, the robot learned a sequence of movements—from closing its gripper to pouring water and stirring—allowing it to eventually perform the task autonomously. Even more impressively, the sensor’s gentle responsiveness enabled a robot to handle a delicate object, such as an egg, without causing damage.

The implications of these achievements are far-reaching. In robotics, the integration of such tactile sensors could lead to machines that are far more responsive and adaptive, capable of executing complex tasks with the care and precision of a human. In healthcare, devices such as smart prosthetics or knee braces could benefit from real-time force feedback, enhancing both safety and functionality. Moreover, the sensor’s ease of fabrication and calibration paves the way for its adoption in everyday applications, making advanced tactile sensing accessible beyond laboratory settings.

Looking ahead, the team plans to refine the sensor further by optimizing its materials and integrating it into more sophisticated systems—including humanoid robots. This continued innovation promises to bridge the gap between human dexterity and robotic precision, ultimately transforming how machines interact with their environments.