Researchers have developed a method for creating musculoskeletal robots using programmable lattice structures that combine soft and rigid mechanics within a single material, enabling lifelike movement and adaptability.
Study: Lattice structure musculoskeletal robots: Harnessing programmable geometric topology and anisotropy. Image Credit: Andrea Izzotti/Shutterstock.com
Published in Science Advances, the study introduces a new approach to robot design by fine-tuning the mechanical behavior of lattice metamaterials through geometric topology and anisotropy control. Using this method, the team built a tendon-driven robotic elephant from a single elastic material, seamlessly blending soft and stiff components to mimic the complex biomechanics found in nature.
Their findings show how bioinspired lattice geometries can lead to lightweight, flexible robots capable of precise, dynamic motion.
Nature as the Blueprint
In animals, movement arises from the interplay of bones, muscles, tendons, and ligaments. This combination allows for power, flexibility, and fine control—whether it’s a cheetah sprinting or a human picking up a fragile object. However, researchers have long struggled to recreate this in robots.
Conventional techniques like multimaterial 3D printing can mimic biological tissues to a degree, but they typically offer only fixed stiffness levels and limited directional control. They fall short of the smooth gradients and dynamic range found in nature.
The team’s solution was to focus on structuring a single material in programmable ways rather than mixing materials.
A Programmable Lattice Design
The researchers created a lattice framework composed of tiny unit cells that can be individually adjusted. Two base geometries—body-centered cubic (bcc) and X-cube—serve as the building blocks. By blending these via a “topology index” ranging from 0 to 1, they could gradually transition between soft and stiff configurations. This method, known as Topology Regulation (TR), allows the robot’s mechanical properties to change smoothly across its structure.
A second method, Superposition Programming (SP), adds another layer of control by rotating, stacking, or translating individual cells. This creates localized zones with tailored stiffness, particularly useful for complex joints.
The entire structure was fabricated from a single elastic resin using high-resolution 3D printing. Mechanical testing confirmed that TR and SP together could recreate the full range of behaviors seen in biological musculoskeletal systems, from rigid support to flexible bending.
The Robotic Elephant: A Functional Demonstration
To test the system, the team built a robotic elephant using only one material throughout. The robot combined soft, flexible sections and strong, load-bearing parts—all achieved through geometric control rather than different materials.
The Trunk
Engineered in three segments (twisting, bending, and helical), the trunk was designed with gradually changing stiffness along its length. This allowed it to perform twisting motions up to 78.1° and bends of 69.6° using just four motors. It could pick up delicate objects like flower stems and lift items more than three times its weight.
The Legs
Each leg had powered hip and knee joints and a passive ankle. Using SP, the joints were fine-tuned to replicate the direction-specific stiffness of real limbs. With tendon-driven actuation, the robot supported over 4 kg (twice its weight) and walked stably with a step length of 150 mm at a speed of 7.5 mm/second.
Broader Potential in Robotics
This programmable lattice approach has far-reaching implications for robotics. Because the mechanical behavior is built into the structure, the need for complex actuation or multimaterial fabrication is reduced. Potential applications include:
- Soft robotic grippers for delicate tasks
- Exoskeletons and prosthetics with more natural motion
- Legged robots capable of navigating uneven terrain
The lattice's open architecture also allows for embedded sensors or adaptive components, enabling smarter, more responsive robotic systems.
What’s Next?
The robotic elephant offers a compelling proof-of-concept for programmable lattice-based design. Future work may explore using different base materials, finer lattice geometries, or integrating real-time feedback systems to enhance adaptability.
By combining flexibility and strength in a single, printable material, this method offers a streamlined path toward robotic systems that move more like living organisms without the complexity of traditional fabrication.
Journal References
Guan, Q., & et al. (2025, 16 July). Lattice structure musculoskeletal robots: Harnessing programmable geometric topology and anisotropy. ScienceAdvances, 11 (29). DOI: 10.1126/sciadv.adu9856, https://www.science.org/doi/10.1126/sciadv.adu9856
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.