SNU Engineers Develop Shape-Shifting Soft Robot for Crawling, Climbing, and Adaptive Movement

Tokyo, Japan (SPX) Feb 18, 2025
Researchers at Seoul National University (SNU) have unveiled a groundbreaking soft robot capable of crawling like a worm, scaling cables, and swiftly altering its shape to shift directions-all powered by a single air input. This innovation presents a new paradigm in soft robotics, enabling enhanced adaptability in various environments.

Redefining Soft Robot Locomotion

Soft robots, constructed from pliable materials, offer flexibility but often require intricate multi-input systems to execute different motions. The new SNU design overcomes these limitations using a snap-through mechanism akin to a toy popper flipping inside out or a Venus flytrap snapping shut. The researchers developed a Snap Inflatable Modular Metastructure (SIMM) that allows the robot to deform smoothly or rapidly transition into a new shape using a single air source.

Versatile Designs for Complex Environments

To showcase the capabilities of this innovative system, the research team built two prototype robots:

- Crawling and Climbing Robot: This robot propels itself forward in a worm-like manner by expanding and contracting its body. Unlike conventional soft robots, it can also grip and climb cables, making it well-suited for maneuvering through challenging terrains.

- Self-Reconfiguring Robot: This model moves forward through bending, but with a sudden snap, it expands into a larger shape, allowing it to shift directions dynamically. This marks the first time a soft robot has demonstrated the ability to change movement paths on demand.

Expanding the Potential of Adaptive Robotics

With its unique crawling, climbing, and shape-shifting capabilities, this technology could pave the way for a range of applications, including search-and-rescue robots that traverse debris, medical devices that navigate the human body, and deployable structures capable of morphing in real time.

Research Report:Snap Inflatable Modular Metastructures for Multi-Path, Multi-Mode Morphing Machines