Energyless robot dances illuminate a new kind of order in an active case

Predicting when and how clusters of grains, robots, or animals will come into order remains a challenge across science and engineering.

In the 19th century, scientists and engineers developed the control of statistical mechanics, which predict how groups of simple grains move between order and disorder, as when a collection of atoms collides. randomly freezes to form a uniform crystal surface.

It is more challenging to predict the general behaviors that can be achieved when the grains become more complex, so that they can move under their own power. This type of system – observed in bird herds, bacterial colonies and robot swarms – goes by the name “active subject”.

As reported in the January 1, 2021 issue of the journal Science, a team of physicists and engineers has proposed a new principle by which active subject systems can be ordered in a fun way, without the need for higher level management or even programmed interaction among the producers. And they have reflected this principle in several systems, including groups of robots that change shape from time to time called “smarticles” – smart, active particles.

The theory, developed by Dr. Pavel Chvykov at the Massachusetts Institute of Technology while a student with Professor Jeremy England, who is now a researcher in the School of Physics at the Georgia Institute of Technology, argues that certain types of active matter with deception are enough. dynamics independently detects what the researchers refer to as “low rattling” states.

“Fighting is when a business takes energy flowing into it and turns it into a random movement,” England said. “Retreat can be greater either when the movement is more violent, or more random. On the other hand, low rattling is either very small or very organized – or both. So , the idea is that if your subject and energy source allow you to rattling low state, the system will randomly reset until it finds that state and then engage it. you power through forces with a particular pattern, this means that the chosen state can find a way to move the case that matches a very pattern. “

To develop their theory, England and Chvykov were inspired by a phenomenon – known as blackmail – discovered by the Swiss physicist Charles Soret in the late 19th century. In Soret experiments, he found that salt solution initially in a tube to a difference in temperature would lead to an increase in salt density in the colder region – corresponding to an increase in the order of solution.

Chvykov and England developed several mathematical models to reflect the low rattling principle, but could not confirm their predictions until they connected with Daniel Goldman, Professor of Dunn Family Physics at the Georgia Institute of Technology.

Goldman said, “A few years ago, I saw England give a seminar and I thought that some of our smarticle robots could be valuable in proving this theory.” Working with Chvykov , who visited the laboratory of Goldman, Ph.D. students William Savoie and Akash Vardhan used three smartphones enclosed in a ring to compare experiments and theory. The students maintained that instead of demonstrating complex dynamics and fully examining the coffin, the robots would organize themselves in a few dances. – for example, one dance involves three robots dragging each other’s arms in sequence. These dances may continue for hundreds of flaps, but suddenly lose stability and are replaced by a dance of a different pattern.

After first showing that these simple dances were very low-cost states, Chvykov worked with engineers at Northwestern University, Professor Todd Murphey and a Ph.D. student Thomas Berrueta, who developed smarter and better controlled smarticles. The improved smarticles allowed the researchers to test the limits of the theory, including how the types and number of dances varied for different arm flapping patterns, as well as how the controls could be controlled. those dances. “By controlling layers of low rattling states, we were able to force the system to make adjustments that do useful work,” Berrueta said. Northwestern University researchers say these findings may have a profound effect on microrobotic separations, active substances, and metamaterials.

As England put it: “For robot swarms, it’s about getting a lot of adaptive and smart body behavior that you can design to implement in one swarm, even if the robots are individual. relatively cheap and relatively simple.For living cells and novel materials, it ‘s probably about understanding what the’ swarm ‘of atoms or proteins will get you, as far as new materials or buildings computing. “

###

Georgia Tech’s team based on the study includes Jeremy L. England, a Living Systems Physics scientist who will study with the School of Physics, Dunn Family Professor Daniel Goldman, Kurt Wiesenfeld professor, and graduate students Akash Vardhan (Speech Biology) and William Savoie (School of Physics). They will be joined by graduate student Pavel Chvykov (Massachusetts Institute of Technology), along with 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 Investigation Office under awards from ARO W911NF-18-1-0101, ARO MURI Award W911NF-19-1-0233, ARO W911NF-13-1-0347, by National Science Foundation under the auspices of PoLS-0957659, PHY-1205878, PHY-1205878, PHY-1205878, and DMR-1551095, NSF CBET-1637764, with James S. McDonnell Foundation Scholar Award 220020476, and Dunn Institute of Technology Family Professor Georgia. The opinions, conclusions, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the organizations that support them. .