Spontaneous robotic dances highlight a new type of order in the active matter

IMAGE: When a swarm of smarticles is made to interact in a confined space, they form amazingly symmetrical dances whose choreography spontaneously emerges from the physics of low noise. view More

Credit: Thomas A. Berrueta

Predicting when and how collections of particles, robots or animals become orderly remains a challenge in science and engineering.

In the 19th century, scientists and engineers developed the discipline of statistical mechanics, which predicts how groups of simple particles transition between order and disorder, as when a collection of randomly colliding atoms freezes to form a network. uniform crystalline.

More difficult to predict are the collective behaviors that can be achieved when particles become more complicated so that they can move under their own power. This type of system – observed in poultry flocks, bacterial colonies and swarms of robots – is called “active matter”.

As reported in the January 1, 2021 issue of the magazine Science, a team of physicists and engineers have proposed a new principle by which active matter systems can order spontaneously, without the need for higher-level instructions or even programmed interaction between agents. And they have demonstrated this principle in a variety of systems, including groups of robots that periodically change shape called “smarticles” – intelligent, active particles.

The theory, developed by Dr. Pavel Chvykov at the Massachusetts Institute of Technology while a student of Prof. Jeremy England, who is now a researcher at the Georgia Institute of Technology School of Physics, argues that certain types of active matter are sufficient disordered. the dynamics will spontaneously find what researchers are referring to “low-noise” states.

“Rattling is when matter takes energy that flows into it and turns it into random motion,” said England. “The rattle can be louder either when the movement is more violent or more random. Instead, the low noise is either very light or very organized – or both. So the idea is that if matter and energy source allow the possibility of a low noise state, the system will randomly rearrange itself until it finds that state and then get stuck in. If you provide energy by force with a certain pattern, this means that the selected state will discover a way to move the material that fits fine with that model. “

To develop their theory, England and Chvykov were inspired by a phenomenon – dubbed dubbed – discovered by the Swiss physicist Charles Soret in the late nineteenth century. In Soret’s experiments, he found that subjecting an initial uniform solution of salt in a tube to a temperature difference would spontaneously lead to an increase in the salt concentration in the colder region – which corresponds to an increase in the order of solution.

Chvykov and England developed numerous mathematical models to demonstrate the principle of low noise, but only after connecting with Daniel Goldman, a physics professor at the Dunn family at the Georgia Institute of Technology, were they able to test their predictions.

Goldman said: “A few years ago, I saw England holding a seminar and I thought some of our smart robots might prove valuable in testing this theory.” Working with Chvykov, who visited Goldman’s doctoral laboratory. students William Savoie and Akash Vardhan used three flapping smarticles enclosed in a ring to compare the experiments with the theory. The students noticed that instead of displaying complicated dynamics and fully exploring the container, the robots would spontaneously organize themselves into a few dances – for example, a dance consists of three robots slapping each other in the arms. These dances could persist for hundreds of claps, but they can suddenly lose their stability and can be replaced by a dance with a different pattern.

After proving for the first time that these simple dances were indeed noisy states, Chvykov worked with engineers at Northwestern University, Prof. Todd Murphey and a PhD. student Thomas Berrueta, who developed more refined and better controlled smarticles. The improved articles allowed the researchers to test the limits of the theory, including how the types and number of dances varied for different patterns of arm fluttering, and how these dances could be controlled. “By controlling the sequences of low noise states, we managed to make the system reach configurations that work usefully,” said Berrueta. Researchers at Northwestern University say the findings could have far-reaching practical implications for microrobotic swarms, active matter and metamaterials.

As England noted: “For swarms of robots, it’s about getting a lot of adaptive and intelligent group behaviors that you can design to achieve in a single swarm, even if individual robots are relatively cheap and simple in terms of view of the calculation. For living cells and new materials, it could be about understanding what the “swarm” of atoms or proteins can bring you, in terms of new materials or computational properties. “


The Georgia Tech team of the study includes Jeremy L. England, a scientist in living systems physics who researches with the School of Physics, Professor Dunn Family Daniel Goldman, Professor Kurt Wiesenfeld and graduate students Akash Vardhan (Quantitative Biosciences) and William Savoie (School of Physics). of Physics). He joins student Pavel Chvykov (Massachusetts Institute of Technology), Professor Todd D. Murphey, and graduate students Thomas A. Berrueta and Alexander Samland of Northwestern University.

This material is based on work supported by the Army Research Office in the ARO W911NF-18-1-0101, ARO MURI Award W911NF-19-1-0233, ARO W911NF-13-1-0347, by the National Science Foundation. pursuant to PoLS-0957659, PHY-1205878, PHY-1205878, PHY-1205878 and DMR-1551095, NSF CBET-1637764, by James S. McDonnell Foundation Scholar Grant 220020476 and Georgia Institute of Technology Dunn Family Professorship. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the sponsoring agencies.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! through contributing institutions or for the use of any information through the EurekAlert system.