Researchers Develop Extremely Dexterous Microrobot

Researchers at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Harvard Wyss Institute for Biologically Inspired Engineering have developed an extremely dexterous microrobot. HAMR-JR is a half-scale version of the previously created cockroach-inspired Harvard Ambulatory Microrobot (HAMR). 

The HAMR-JR is just the size of a penny, but it is capable of performing all of the tasks of the larger HAMR. It is one of the most dexterous microbots to be created. 

The research titled “Scaling down an insect-size microrobot, HAMR-VI into HAMR-Jr” was presented virtually at the International Conference on Robotics and Automation (ICRA 2020).

Kaushik Jayaram is a former postdoctoral fellow at SEAS and Wyss, as well as the first author of the paper. Jayaram is also an Assistant Professor at the University of Colorado, Boulder.

“Most robots at this scale are pretty simple and only demonstrate basic mobility,” said Jayaram. “We have shown that you don't have to compromise dexterity or control for size.“

One of the tasks going into the project was to figure out whether the pop-up manufacturing process, which was used to build previous versions of HAMR and microbots like the RoboBee, could be used to construct robots at multiple scales. This could be as small as surgical bots or as large as industrial robots. 

PC-MEMS

The process used to build HAMR-JR is called printed circuit microelectromechanical systems, or PC-MEMS. In this process, the robot’s components are put into a 3D structure after being etched into a 2D sheet. For the HAMR-JR, the 2D sheet design of the robot, actuators, and onboard circuitry were shrunk down to create a smaller robot with the same functionalities.

“The wonderful part about this exercise is that we did not have to change anything about the previous design,” said Jayaram. “We proved that this process can be applied to basically any device at a variety of sizes.”

Scaling Down HAMR-JR

HAMR-JR is only 2.25 centimeters in body length with a weight of 0.3 grams. On top of being one of the smallest microrobots, it is also one of the fastest, capable of moving 14 body lengths per second. 

Principles such as stride length and joint stiffness do get affected when the robot is scaled down. To get around this, the researchers also developed a model that is able to predict locomotion metrics such as running speeds, foot forces, and payload depending on target size. With this model, a system can be developed with the right specifications. 

Robert Wood is co-author of the paper and a Charles River Professor of Engineering and Applied Sciences in SEAS. He is also a Core Faculty Member of the Wyss. 

“This new robot demonstrates that we have a good grasp on the theoretical and practical aspects of scaling down complex robots using our folding-based assembly approach,” Wood said.

The research was supported by DARPA and the Wyss Institute. 

Other co-authors included Jennifer Shum, Samantha Castellanos, and E. Farrell Helbling.