Picture a robot; make it smaller; make it smaller still. What do you end up with? If your answer is something that looks a bit like an insect, you’re pretty close. Picotaur, a new robot from the labs of Sarah Bergbreiter and Aaron Johnson at Carnegie Mellon University measures less than 1 centimeter, but it can run, turn, push loads and climb miniature stairs. This tiny machine could transform how we approach search and rescue, inspection, and exploration in areas too small for larger robots to access.

Tiny robot legs

Creating legged robots capable of navigating rough terrains and overcoming obstacles has always been a challenge. Larger robots can easily be equipped with complex mechanisms that allow them to move in three dimensions, but replicating this on a smaller scale is significantly more challenging. Insect-scale robots, defined as those weighing less than a gram, have historically struggled to achieve the same level of agility and versatility.

Herein lies the main innovation of the Picotaur — its legs.

“This robot has legs that are driven by multiple actuators so it can achieve various locomotion capabilities,” said Sukjun Kim, a recent Ph.D. graduate advised by Bergbreiter. “With multiple gait patterns it can walk like other hexapod robots, similar to how a cockroach moves, but it can also hop from the ground to overcome obstacles.”  

The robot measures just 7.9 millimeters and was 3D printed using a process called “two-photon polymerization,” one of the most precise 3D-printing techniques. Previously, this approach was used to build various small-scale robotic systems in the lab such as microbots, microgrippers, microswimmers and microsensors.

“Using this process, we were able to miniaturize the two-degree-of-freedom linkage mechanism that allows Picotaur to clear step heights and easily alternate between walking and jumping,” said Bergbreiter, a professor of mechanical engineering. A two-degree-of-freedom linkage mechanism is a mechanical system that allows movement in two independent directions, enabling complex motions such as rotation and translation.

Each leg is powered by rotary microactuators, which drive the robot using voltage inputs. These actuators allow Picotaur to walk using a variety of gait patterns, including an alternating tripod gait and a pronk gait. This alternating tripod gait enables the robot to move swiftly across flat surfaces, while the pronk gait allows it to hop, providing the necessary ground clearance to navigate more complex terrains.

Small but mobile

The Picotaur is also fast. It can reach speeds of up to 57 mm per second, which is equivalent to 7.2 times its body length per second — an impressive feat for a robot so small.

The researchers tested the Picotaur’s ability to push objects on a miniature soccer field. The robot pushed a small payload (a 4 mg foam ball) to a goal, reorienting and adjusting itself in the process.

There’s still a way to go before Picotaur can start working in the “wild”. Currently, the robot relies on external wires for power and control, which is a big limitation. Researchers are working on installing lightweight batteries or solar cells that would allow the robot to operate untethered.

Nonetheless, Picotaur represents a significant step forward in the field of microrobotics. Its combination of advanced manufacturing techniques, innovative leg mechanisms, and versatile locomotion capabilities make it a promising candidate for a wide range of applications. As researchers continue to refine its design and capabilities, we can expect to see even more impressive feats from this tiny robot.

“Historically, microfabrication technology was limited in manufacturing microscale devices in two-dimensional spaces, like for the semiconductor industry,” said Kim. “But now we have this capability to expand the design space from 2D to 3D. We can apply this process to create other small-scale robotic systems for various applications, for example, microgrippers for grasping and delivering small objects for surgical applications and microscale manufacturing applications.”

The study was published in the journal Advanced Intelligent Systems.