Engineers at the University of Texas at Austin have developed a robotic hand capable of handling some of the most fragile objects imaginable—including potato chips and raspberries—without breaking them. The system achieved a 91.9% success rate in single-trial grasping experiments, demonstrating a major step forward in robotic touch and precision.
Teaching Robots a Gentle Touch
Robots have long been excellent at lifting heavy objects and performing repetitive industrial tasks. However, handling delicate items remains a major challenge. Many robotic grippers rely heavily on cameras and pre-programmed motions, which often causes them to apply too much pressure.
To address this issue, researchers created a system called FORTE (Fragile Object Grasping with Tactile Sensing). This technology combines soft robotic fingers with built-in sensors that allow the robot to detect how much force it is applying. As the robotic fingers make contact with an object, the sensors quickly measure pressure changes and help the system adjust its grip in real time. As a result, the robot can hold fragile items without crushing them.
Inspired by the Structure of Fish Fins
The design of the robotic fingers draws inspiration from nature, particularly the fin-ray effect found in fish fins. Using this concept, the team 3D-printed flexible fingers with tiny internal air channels. When the robot grasps an object, these channels shift slightly and create small pressure variations.
Sensors inside the fingers detect these pressure changes and immediately signal the system to modify the gripping force. Consequently, the robotic hand can maintain a secure yet gentle hold on objects that would normally be difficult for machines to handle.
Promising Results and Future Uses
To test the system, researchers asked the robot to pick up 31 different objects with varying shapes, textures, and fragility. These included items such as raspberries, potato chips, apples, soup cans, and billiard balls. Impressively, the robot successfully grasped objects on the first attempt 91.9% of the time. It also detected most slipping events with very high accuracy.
In the future, robots equipped with this type of tactile sensing could handle delicate foods in packaging facilities, assemble fragile electronic components, or assist with precise medical procedures. Ultimately, the technology brings robots closer to matching the subtle sense of touch that humans use every day.
