This Soft Robot Can Dress a Moving Person in About 10 Seconds
A KAIST and Stanford team has demonstrated a garment with air-powered, vine-like soft robots that unfurl clothing along the wearer's body without hands or external assistance. The prototype works while the person moves and may support people with limited mobility, emergency workers and cleanroom staff, but it still depends on pneumatic hardware and has not been validated for everyday independent use.
Clothing that moves onto the body
Robotic dressing systems usually try to manipulate ordinary clothes with rigid mechanical arms. That creates a difficult problem: fabric folds unpredictably, the human body moves, and a hard robot must operate close to vulnerable skin.
Researchers from KAIST and Stanford University have demonstrated a very different approach. Instead of building a robot that handles the garment, they turned the garment itself into a soft robot that can unfurl around the wearer.
The prototype, called a Self-Wearing Adaptive Garment, can put a full suit onto a person in about 10 seconds, according to the research team. The wearer does not need to use their hands, receive help from another person or remain perfectly still.
How a vine becomes a dressing robot
The design borrows a movement strategy from climbing plants. Soft tubes are folded inside themselves and embedded into the clothing. When air pressure is applied, each tube grows from its tip by turning itself outward, rather than dragging its entire length across the body.
As the tubes extend, they carry and unfurl the fabric close to the person's shape. Because only the growing tip moves relative to the surface, the system reduces friction between clothing and skin. It can follow curves and continue across surfaces that are slippery, sticky or sloped.
This is an example of embodied intelligence. Much of the useful behaviour comes from the mechanics and geometry of the soft material, so the system does not require a complicated AI controller to calculate every fold and body movement.
Why it matters beyond convenience
Dressing is a basic daily activity that can become difficult after a spinal injury, stroke, amputation or loss of strength and coordination. Existing assistive robots are often expensive, slow, rigid and dependent on a carefully controlled posture.
A garment that can deploy itself could eventually give some users more privacy and independence. The researchers also see potential in environments where people must dress quickly without contaminating the clothing, including semiconductor cleanrooms, medical facilities and emergency response.
Protective equipment is another possible use. A firefighter, hazardous-material worker or cyclist could theoretically activate clothing while keeping both hands occupied. That vision remains ahead of the current prototype, but it explains why speed and operation during movement are important.
What the demonstration does not prove
The work is a research prototype, not a ready-to-buy garment. It relies on an external source of compressed air and associated hardware. A practical product would need a compact, quiet and safe pressure system that adds little weight.
Real clothing also varies in size, stiffness, closures and layering. The system must be tested across many body shapes, mobility limitations and uncontrolled home environments. It also needs reliable ways to stop immediately if fabric catches, pressure becomes uncomfortable or the wearer loses balance.
The published paper appeared in the January 2026 issue of IEEE Robotics and Automation Letters, while the team publicly demonstrated the technology in South Korea in July. The demonstration gives the work a vivid real-world form, but it does not replace larger usability and safety studies.
A quieter direction for robotics
The project is notable because it does not depend on a humanoid machine or a powerful general-purpose AI model. It solves a human problem by making the material itself adaptive.
If the pneumatic equipment can be miniaturised and the garment can pass safety testing, soft robotic clothing could become a practical assistive technology. For now, its strongest achievement is showing that getting dressed can be treated not as a robotic manipulation problem, but as a carefully designed movement built into the fabric.
Sources and citations
- IEEE Robotics and Automation Letters: Self-Wearing Adaptive Garments via Soft Robotic Unfurling
- KAIST research record: Self-Wearing Adaptive Garments via Soft Robotic Unfurling
- Reuters: South Korea-US team unveils robotic technology that dresses the wearer
- NSF Public Access Repository: Self-Wearing Adaptive Garments
Published by
NewTqnia Editorial
Technology & innovation desk