This Robotic Gripper Can Sense Ripeness by Touch

Robots are a big help on modern farms, where they can do everything from watering crops to bringing in the harvest. But while they are very capable in many respects, they don’t have the same “feel” for things that people do. Sure, they can pluck countless bushels of fruit from an orchard, but they aren’t so good at telling whether that fruit is any good. That takes a special kind of knowledge. And if you’ve ever seen anyone in a grocery store poking and prodding at produce, you know just how involved the selection process can be.

In the future, robots may give us a run for our money, thanks to the work of a team of researchers at Cornell University. They have developed a Raspberry Pi-powered soft robotic gripper capable of sensing the ripeness of fruit through touch, mimicking the nuanced way humans assess produce.

The research centers on a five-fingered, flexible gripper embedded with stretchable fiber-optic sensors. Unlike traditional rigid robotic claws, this design allows the machine to gently grasp delicate fruits like strawberries without bruising them. At the same time, it measures subtle differences in firmness — one of the most reliable indicators of ripeness.

The project combines multiple sensing methods into a single compact device. Each finger contains optical fibers that detect both curvature and pressure by measuring changes in light intensity as the material bends or presses against an object. Meanwhile, a small Raspberry Pi Camera Module embedded in the palm provides visual data, helping the robot identify fruit that may be hidden behind leaves.

This multimodal approach — blending touch, vision, and motion sensing — addresses a major limitation in current agricultural robots. Many existing systems rely heavily on visual cues like color, which can be unreliable due to lighting conditions or natural variation between fruits. Others use tactile sensors but lack the finesse needed to avoid damaging soft produce. By integrating these capabilities, the Cornell gripper can make more accurate decisions while handling crops more delicately.

The device is not only intelligent but also efficient. It can close its grip in under two seconds, lift objects weighing up to one kilogram, and adapt to a wide range of shapes and sizes. Once it has determined that a fruit is ready to harvest, a built-in rotational mechanism gently twists it from the vine instead of pulling, reducing strain on both the fruit and the plant.

While strawberries served as the primary testing model — since their color provides an easy way to verify ripeness — the technology could be even more valuable for fruits like avocados or pawpaws, where visual cues are less reliable. These crops often suffer high postharvest losses because they ripen unpredictably and bruise easily.

Beyond improving harvesting accuracy, the researchers believe their system could support more sustainable farming practices. Smaller, more intelligent robots could enable diverse planting strategies, reduce reliance on pesticides, and minimize food waste in the years ahead.