Robotic Hand Achieves Delicate Touch, Capable of Handling Fragile Objects

AUSTIN, TEXAS – Researchers at the University of Texas at Austin have developed a robotic hand with tactile sensing capabilities sensitive enough to grasp a potato chip or a raspberry without causing damage. The new technology, dubbed Fragile Object Grasping with Tactile Sensing (FORTE), represents a significant step forward in robotics, potentially impacting industries from food processing to healthcare.

The breakthrough centers on combining advanced tactile sensing with soft robotics, addressing a key limitation in current robotic systems. “Right now, robotics is starting to be able to do large motions around the house, but struggles with really fine and delicate movements,” explained Siqi Shang, a doctoral student at UT Austin’s Cockrell School of Engineering and lead author of a paper detailing the research, published in IEEE Robotics and Automation Letters. “Robots can fold a shirt but may struggle to carefully pick up your glasses or unpack fruit from your groceries. We believe sensing signals will give robots a sense of touch to handle these objects carefully.”

The robotic fingers are inspired by the “fin-ray effect,” a design principle found in the structure of fish fins. Utilizing advanced 3D-printing techniques, the fingers incorporate internal air channels that function as tactile sensors. These channels shift when the fingers prepare to grasp an object, creating changes in air pressure. Small, commercially available sensors detect these pressure changes, providing the robot with real-time force feedback and alerting it to any slippage.

Researchers tested the grippers on a diverse set of 31 objects, including fragile items like raspberries and potato chips, slippery items such as jam jars and billiard balls, and common household objects like soup cans and apples. The system demonstrated a 91.9% success rate in single-trial grasping experiments, surpassing the performance of traditional grippers that rely solely on visual feedback. Crucially, the system accurately identified 93% of slips with 100% precision, meaning it never incorrectly signaled a slip event.

“Humans pick up objects with just the right amount of force; too much and you’ll crush it, but too little and it’ll slip out of your hand,” said Lillian Chin, assistant professor of electrical and computer engineering at UT Austin. “Most current force sensors aren’t fast or accurate enough to provide that Goldilocks level of detail. In particular, our sensors operate closer to the timescales of human hand sensors.”

Beyond speed and accuracy, the 3D-printed sensors offer durability and customization. The researchers noted the ease with which the sensors can be adapted to various shapes. The slip-sensing capability is a particularly distinguishing feature, as few existing robotic gripping technologies incorporate such a function, and those that do lack FORTE’s speed and responsiveness.

Potential applications for FORTE are wide-ranging. In the food processing industry, the technology could minimize waste and improve efficiency when handling delicate produce and baked goods. In healthcare, robots equipped with FORTE could precisely manipulate medical instruments or fragile biological samples. Manufacturing could also benefit, with the technology enabling the handling of delicate components like electronics and glassware. The researchers collaborated with the College of Fine Arts’ theatre and dance department for a year, applying the sensing technology in that context.

To facilitate further development, the research team has made the hardware designs and algorithms publicly available. Ongoing operate focuses on mitigating the sensors’ sensitivity to temperature fluctuations and enhancing their ability to recover from slipping objects. The research was supported by the Texas Robotics Industrial Affiliate Program, the National Science Foundation, the Office of Naval Research, the DARPA TIAMAT program, and South Korea’s Institute of Information & Communications Technology Planning & Evaluation.