In northeastern Brazil, termites have transformed an area about the size of Great Britain by sensing their environment, communicating with each other about it and working together to build vast tunnel systems. Inspired by this and other instances of the small but mighty collective activity in nature, Cornell physics researchers have made robots that for the first time can sense the temperature of their surroundings and react together to change it.
“Little things can have a large impact,” said Itai Cohen, the Josephson Family Professor of physics in the College of Arts and Sciences (A&S), a leader in developing microscopic robots and a corresponding author on the study.
This is the first example of microscopic robots that can alter their physical environment, pumping liquid from hot regions to cold regions or vice versa. Three key capabilities – environmental sensing, fluid manipulation through artificial cilia and communication among microchips – create a feedback loop inspired by systems in nature that could be used in future medical or agricultural applications.
These robots combine previous robots’ skill sets. In recent studies, Cohen and his lab have developed robots a hair’s-breadth in diameter that walk autonomously, sense their environment, make images and measurements, and communicate with each other.
“We’re building out the repertoire of things these robots can do,” Cohen said. “The idea here is, how do we get them to work together to achieve big manipulations of their environment, and this is a first step.”
“Microscopic Robots that Sense and Reshape Their Environment” published in Nature Electronics on Sept. 23. Postdoctoral researcher Jinsong Zhang and former doctoral student Wei Wang, Ph.D. ’24, from Cohen’s lab are co-first authors. Zhang and Wang are corresponding authors.
“Nature’s organisms can give us some really interesting things to think about,” Zhang said. Designing this robot, he looked to the cilia that the paramecium, a single-celled water creature, uses to move, feed itself and pump fluids – including sensing and eradicating hazardous chemicals.
The robot “cilia” are inspired by the fluid-pumping principle of natural cilia, Zhang said. But rather than directly replicating their structure, the researchers designed a more efficient two-hinge architecture that can be readily programmed by electronic signals. Sequential movement of the two hinges produces a nonreciprocal, paddle-like motion that drives the surrounding liquid.
One paddle isn’t enough to make a difference, so many need to work together. The team set 54 hinged cilia in an array equipped with two temperature sensing circuits, a leader and a follower, which communicate with each other and synchronize the cilia in the array to pump together in response to the temperature. In this electronic component, the amount of electronic current depends on the temperature; it was developed by co-author Alyssa Apsel, the IBM Professor of Engineering in the School of Electrical and Computer Engineering, in the Cornell Duffield College of Engineering.
“These circuits essentially tell the cilia array which hinge should pump when,” Cohen said. “If the temperature is above a certain point, then it says pump in the forward direction. If it’s a lower temperature, it says pump in the reverse direction.”
All the cilia in the array synchronize, a behavior related to research Cornell mathematician Professor Steven Strogatz (A&S) has done on synchronization of coupled oscillators – that is, how things in nature tend to sync up.
“The same thing is happening here. It’s a classic synchronization mechanism,” Cohen said. “Only this time it’s happening electronically. The chips are sending little voltage pulses to each other and through that they’re able to synchronize their frequency exactly. The one that’s fastest becomes the leader. The one that’s slowest lags by some phase – a quarter cycle. This strategy was inspired by fireflies.”
In future iterations of the concept, robots could respond to factors like light or pH level, setting off chemical reactions or mechanical motions, Cohen said. Instead of being set in a static array, the interacting robots could walk independently, programmed to react in a specific way to certain environmental cues, such as in medical or agricultural settings.
“The ability to pay attention to the environment is important,” Cohen said. “That’s the promise of this: collective behaviors that emerge from local interactions but are also governed by the environment in which these robots are working.”
This project was primarily supported by the Army Research Office and the U.S. National Science Foundation. Partial support was also provided by the Kavli Institute at Cornell for Nanoscale Science. This work was performed in part at the Cornell NanoScale Science and Technology Facility.
Cornell co-authors include postdoctoral researcher Yanxin Ji, Ph.D. ’23; former postdoctoral researcher Zexi Liang and Itay Griniasty; doctoral students Paragkumar Chaudhari; Kyubum Shim, M.S. ’25; Chibin Zheng, M.S. ’26; and Jacob Pelster, Ph.D. ’26. Additional co-authors include John Severn and Eric Lauga from the University of Cambridge; Daniel Seara from University of Illinois, Chicago; and Vincenzo Vitelli from University of Chicago.
Read the story in the Cornell Chronicle.
More News from A&S
