A New System to Control Swarm of Tiny Robots is Onboard!

Employing vibration instead of onboard sensors to command a horde of Microbots.

The small scale of microrobots makes them perfect for medicine administration, disease diagnostics, and even surgery. Individual robots can work together as swarms to make significant breakthroughs in everything from building to surveillance. Due to their small size, microrobots frequently have limited sensing, communication, motility, and processing capabilities despite their potential; nevertheless, a new study from the Georgia Institute of Technology improves the effectiveness of their collaboration. The article proposes a novel control scheme for the controllable aggregation and dispersion of swarms of 300 3-micron microbristle robots (microbots). The innovation is special to Georgia Tech because of its proficiency in robotics, electric and computer engineering, and its drive for multidisciplinary partnerships.

Microelectromechanical systems (MEMS) on silicon and the appearance of the microcontroller in the final decade of the 20th century gave rise to microbots, but many of them just employ silicon for sensors and other mechanical components. The early 1970s classified study for American intelligence agencies resulted in the conceptual design and research for the first miniature robots of this size. Applications like electronic intercept missions and prisoner-of-war rescue support were foreseen at the time. This early set of calculations and concept design did not immediately translate into advances in prototype development because the underlying miniaturization support technologies were not completely matured at the time. The tiniest microrobots as of 2008 use a scratch drive actuator.

Azadeh Ansari, an assistant professor in the School of Electrical and Computer Engineering, claimed that by working with roboticists, they were able to "narrow the gap" between single robot design and swarm control (ECE). So I guess the various components were already present, and we merely connected them. Controlling Collision-Induced Aggregations in a Swarm of Micro Bristle Robots is the title of the study that the researchers published in IEEE Transactions on Robotics.

Microbots lack the capacity to carry the same sensors, communications, or power units as larger robots, which can move by detecting the environment and wirelessly communicating this data to one another. Instead, in this work, the researchers used physical interactions between robots to stimulate robot swarming. "Microbots are too small to interpret and make decisions," said Zhijian Hao, an ECE Ph.D. student. "However, we could influence how individual robots move and the collective behaviors of hundreds and thousands of these tiny robots by using the collision between them and how they respond to frequency and the amplitude of global vibration actuation."

Both the linear motion of microbots and the unpredictability of their rotation are governed by these behaviors or motility characteristics. The researchers were able to conduct motility-induced phase separation and manipulate these motility features using vibration (MIPS). The idea was adapted by the researchers from thermodynamics, which describes how an agitated material can transform between solid, gas, and liquid phases. In order to achieve adequate spatial coverage, the researchers altered the vibration level, causing the microbots to cluster or disperse. They created computational models and a live tracking system for the 300 robot swarm utilizing computer vision to better comprehend these phase separations. These made it possible for the researchers to examine the mobility and behavior data of the microrobots that produce the traits of the swarm.

The interdisciplinary nature of the research is responsible for the project's success. The robotics researchers brought modeling experience, while the ECE researchers had competence in creating microelectromechanical systems (MEMS) to produce technologies such as computer chips or microbots. The partnership with IRIM Director and Professor Seth Hutchinson, Professor Magnus Egerstedt, currently at the University of California, Irvine, and their Ph.D. students Sid Mayya and Gennaro Notomista began when Ansari first developed microbristle bots in 2019 using 3D-printed polymers. Ansari claimed that while they were more knowledgeable about algorithms, modeling, and closed-loop and open-loop control, "we knew more about how to make micro devices and actuate them." Therefore, the multidisciplinary collaboration was excellent since each group benefited from the fresh viewpoints that the others brought to the table.