1.5-Centimeter Biohybrid Robot Jumps 5mm in a Single Contraction: ETH Zurich Unlocks It with Scaffold Design

A noteworthy development has emerged in the field of biohybrid robotics: researchers at ETH Zurich have made a 1.5-centimeter-long microrobot jump 5 millimeters—one-third of its own body length—using just a single contraction of lab-grown muscle. This leap in jumping capability does not come from stronger muscle or greater energy input, but from a seemingly unremarkable design change: the way the muscle connects to the skeleton.

Core Breakthrough: Not Stronger Muscle, but Smarter Connection

The research team's key idea was to design the muscle and the flexible 3D-printed scaffold as a single integrated system, rather than building the robot skeleton first and then "attaching" muscle to it. The scaffold is etched with tiny grooves that serve two purposes: guiding muscle fibers to grow in a specific aligned direction, and providing firm anchor points for muscle contraction.

When the muscle contracts, force is efficiently transmitted to the scaffold rather than lost at the connection point. It is precisely this improvement in "force transmission efficiency" that allows the same single contraction to generate enough reaction force to lift the robot off the ground. In other words, the leap in jumping capability comes from optimizing the mechanical interface, not from a performance breakthrough in the biological material itself.

According to the Reddit source, the researchers achieved a 5-millimeter jump height by changing the way muscle attaches to the frame.

Biohybrid robotics refers to robotic systems that combine living biological tissues (such as muscle cells or neurons) with artificial mechanical structures. Their core advantage lies in leveraging the natural high power density and low energy consumption of biological tissues to achieve actuation performance at the microscale that conventional motors struggle to match. Common biological actuators include cardiomyocytes, skeletal muscle cells, and insect flight muscles. Such robots typically need to solve three major problems: nutrient supply, directed tissue growth, and force transmission efficiency. ETH Zurich's work targets the latter two, optimizing both muscle alignment and the mechanical interface through scaffold design.

Why the "Grooves on the Scaffold" Matter So Much

In biohybrid robots, living muscle cells often grow in a disorganized manner. When fiber directions are inconsistent, forces cancel each other out during contraction, and overall output drops sharply. It's like a rowing crew where everyone rows in a different direction—the boat naturally won't go fast.

The microgrooves on the scaffold solve the alignment problem, ensuring fibers face the same direction so contraction forces add up rather than cancel out. At the same time, the grooves provide a mechanical interlocking structure that prevents the muscle from slipping off the scaffold when generating force. Together, these two effects allow the explosive power of a single contraction to be fully utilized.

From an engineering perspective, the value of this work lies in treating "biological" and "mechanical" as a coordinated system designed together. Traditional biohybrid robotics research often focuses on improving the muscle cells themselves, whereas the ETH Zurich team turned to interface engineering, offering a lower-cost and more reproducible optimization path.

Force transmission efficiency refers to how much of the force generated by an actuator is effectively transmitted to the load or structure. In biohybrid robots, the force generated by muscle contraction must pass through multiple connections—the cell-matrix interface, the matrix-scaffold interface, and others—before acting on the external environment. Slippage, deformation, or stress concentration at these interfaces all cause energy loss. The mechanical interlocking structure provided by microgrooves increases contact area and friction, making the muscle tissue less likely to detach from the scaffold during contraction, thereby converting more contraction force into the robot's kinetic energy. This concept is similar to the tendon-bone attachment structure in the musculoskeletal system, where transition layers such as fibrocartilage enable efficient force transfer.

Why Jumping Is So Hard for Microrobots

For centimeter-scale robots, jumping is an extremely challenging action. The smaller the body, the more limited the available space, energy, and actuation stroke. Conventional motors see efficiency drop sharply at the microscale, whereas biological muscle happens to offer high power density at small scales.

But muscle has its shortcomings too: it needs to be attached to a structure to do work, and the quality of the attachment interface directly determines overall performance. This study demonstrates that even without changing the muscle's own contraction capability, optimizing only the attachment structure can give the robot considerable jumping ability. Five millimeters relative to a 1.5-centimeter body length means this robot has preliminary mobility for scenarios such as crossing obstacles and clearing gaps.

Power density refers to the power output per unit mass or unit volume. At the microscale, conventional electromagnetic motors see power density drop sharply as size decreases due to friction, heat dissipation, and manufacturing precision limitations. In contrast, biological muscle fibers can maintain relatively high power density at the millimeter scale because their energy conversion occurs at the molecular level and does not rely on macroscopic mechanical components. This makes biohybrid actuation an ideal candidate for explosive actions such as jumping and crawling in centimeter-scale robots. However, the power density advantage of muscle can only be realized when force transmission efficiency is sufficiently high; otherwise, most of the energy is lost at the connection interface.

What's Next for Biohybrid Robots

This achievement is still at the laboratory stage, and both jump height and controllability remain far from practical application. But its design philosophy—co-designing muscle and scaffold, using geometric structures to guide biological tissue growth and optimize force transmission—offers a generalizable methodology for microrobots.

Possible future directions include: achieving directionally controllable jumping through multiple muscle groups, adjusting scaffold geometry to change jump height, and integrating such actuators into more complex microrobot systems. For fields requiring centimeter-scale robots, such as search-and-rescue exploration and minimally invasive medicine, this low-energy, high-power-density biohybrid actuation approach is worth continued attention.

According to a single source (Reddit), the core data of this research are a 1.5-centimeter body length and a 5-millimeter jump height; there is currently no cross-validation from other independent sources, and readers may watch for the formal publication of a peer-reviewed paper.

One-line Summary

By optimizing how muscle connects to the scaffold, the ETH Zurich team enabled a 1.5-centimeter biohybrid robot to jump 5 millimeters in a single contraction.

Paragraph Summary

Researchers at ETH Zurich enabled a 1.5-centimeter-long biohybrid robot to jump 5 millimeters—one-third of its body length—using just a single contraction of lab-grown muscle. This breakthrough does not come from stronger muscle or greater energy input, but from the co-design of muscle and a flexible 3D-printed scaffold. Microgrooves on the scaffold guide muscle fibers to align directionally and provide firm anchor points, significantly improving force transmission efficiency. This work treats biological and mechanical elements as a coordinated system, offering a low-cost, reproducible actuation optimization path for centimeter-scale microrobots. The results are still at the laboratory stage and come from a single source, so watch for the formal publication of a peer-reviewed paper.