Collaborative robots face a fundamental sensing problem: they need to detect approaching humans from a safe distance while also being sensitive enough to handle fragile objects without crushing them. Until now, engineers have had to choose. Small sensors give precision but only work up close. Larger ones reach farther but lose the finesse needed for careful manipulation.
Researchers at South China University of Technology have found a way to give robots both abilities at once, using a flexible electronic skin that adapts in real time—much like your pupils adjusting between reading a book and spotting a car on a dark road.
How the Adaptive Sensor Works
The system uses a layer of electrodes covered by a dynamic shield that can electrically expand or contract the sensing field. When the shield narrows, it concentrates the electric field into individual sensor units, letting the robot detect the lightest touch—useful when picking up an egg or threading a needle. When the shield expands, the field stretches farther into space, allowing the robot to sense a human hand approaching from 90 millimeters away, giving it time to stop before contact.
We're a new kind of news feed.
Regular news is designed to drain you. We're a non-profit built to restore you. Every story we publish is scored for impact, progress, and hope.
Start Your News DetoxThe inspiration was biological. "We applied the exact logic of human pupils to electric fields," explained Dr. Xie. Your eye contracts to sharpen fine details and dilates to gather more light in darkness. The sensor does something similar with electromagnetic fields, adjusting its sensing "aperture" based on what matters in the moment.
In tests, the sensor doubled the detection range compared to traditional dual-mode systems—a 104.56 percent increase. That 90-millimeter range might sound small, but in a factory floor where a robotic arm shares space with human workers, it's enough for safe collision avoidance. Once contact happens, the sensor can register forces as light as a few grams while tolerating pressures up to 400 kilopascals, making it suitable for everything from assembly work to handling delicate components.
The breakthrough matters because it collapses what used to require multiple sensors into one adaptive layer. A robot doesn't need separate systems for proximity and touch anymore—one piece of electronic skin handles both. That simplification cuts design complexity, reduces cost, and makes collaborative robots more practical for real manufacturing environments.
The Work Ahead
The team acknowledges real obstacles remain. Creating the microscopic pore structures that give the sensors their sensitivity requires precision manufacturing, and there's currently a 6.3 to 6.8 percent performance variation between individual units. More significant is the challenge of industrial noise: capacitive sensors are vulnerable to electromagnetic interference, temperature swings, and humidity shifts. A factory floor full of heavy machinery could jam the readings.
Future versions may need better electromagnetic shielding or machine learning algorithms that filter out noise in real time. These are engineering problems with known solutions, not fundamental limits—the kind of challenges that typically get solved once a technology proves promising enough to justify the investment.
The findings, published in the International Journal of Extreme Manufacturing, suggest we're moving toward robots that don't just share space with humans, but do so safely and skillfully, using a single adaptive sensor layer instead of a patchwork of different technologies.
