Summary: Researchers at Harvard’s John A. Paulson School of Engineering and Applied Sciences (SEAS) have developed a 3D-printing technique to create artificial muscle-like filaments that bend, twist, and contract when heated. This breakthrough could enable robots with more natural, flexible movement and applications in adaptive grippers, biomedical devices, and shape-shifting structures.

Scientists at Harvard University have unveiled a new 3D-printing method that produces artificial muscles capable of complex, programmed movement, bringing robots one step closer to human-like flexibility.

The approach combines “active” liquid crystal elastomers, which contract when heated, with “passive” elastomers that resist deformation. By printing the two materials side by side and using a rotating nozzle to align the molecules, researchers can program fibers to bend, coil, twist, or expand—without gears, motors, or rigid joints.

In demonstrations, printed soft lattices and wavy filaments transformed in striking ways under heat: flat lattices became dome-shaped, while soft robotic grippers could gently lower onto objects, grasp them, lift, and release—all using only the programmed material properties.

“Robots today move well using motors and hydraulics, but their motions are rigid and mechanical,” said a member of the research team. “Our technique allows for fluid, lifelike movement directly encoded in the material itself.”

Potential applications include adaptive robotic grippers, temperature-responsive structures, biomedical devices, and other systems requiring highly customizable, soft actuators. Because the method is compatible with 3D printing, it could facilitate designs that are difficult or impossible to achieve with conventional motors or pneumatic systems.

Challenges remain. The current system relies on heat to activate the fibers, which limits response time and energy efficiency. Researchers note that while the technology is not yet suitable for high-power robotic applications, it represents a significant step toward more natural, biologically inspired robots.

Harvard SEAS researchers believe this work opens new possibilities for soft robotics, where machines interact with the world in more human-like, adaptable ways.