My Opinion: Swarms are Fascinating and Important

Ed Brown

Recently, my interest has been aroused by research projects involving swarms. How is it that multiple individuals, such as drones or robots, moving independently, without centralized control, can coordinate their actions? One reason I am drawn to this is that I live in busy downtown Manhattan and have often thought about how large numbers of pedestrians walking in different directions manage to negotiate getting to their destinations with hardly a collision.

Aaron Becker, University of Houston Associate Professor of Electrical and Computer Engineering, for example, is interested in methods for controlling swarms of small, low-cost drones. Becker’s team is investigating algorithms that can enable drones to respond dynamically to obstacles, vehicles, predators, and insect swarms, much like birds flying in flocks and fish swimming in schools. “These movements are not pre-programmed but are based on local decisions by individual birds or fish,” said Becker.

Why Swarms?

What’s so special about swarms; why bother? Scientists at Hokkaido University in Japan, explain that, “A swarm of cooperating robots gains a number of characteristics that are not found in individual robots — they can divide a workload, respond to risks, and even create complex structures in response to changes in the environment.” And if there is no centralized control, there is no single point of failure — the task can continue even if one or more individuals break down.

Communicating and Coordinating

Notre Dame Professor and Robotics Engineer Yasemin Ozkan-Aydin built four-legged robots, each with a lithium polymer battery, a microcontroller, and three sensors. Along with a light sensor, two magnetic touch sensors at the front and back of each robot allow the systems to connect to each other. Ozkan-Aydin proposed that a physical connection between robots could enhance mobility to move around on rough terrain and in tight spaces. “You don’t need additional sensors to detect obstacles because the flexibility in the legs helps the robot to move right past them,” she said.

When an individual unit becomes stuck, it sends a light to additional robots to show a request for assistance. Upon sensing the light, the helper robots connect and provide support — a push as they walk together — to successfully traverse obstacles while working collectively.

At Northwestern University, Evanston, Illinois, a team of researchers led by Professor Michael Rubenstein has developed a decentralized algorithm to control swarms of robots. It was successfully tested on a simulation of 1,024 robots and on a swarm of 100 real robots in the laboratory. The robots reliably, safely, and efficiently converged to form a predetermined shape in less than a minute. “Using these multi-robot systems can offer more parallelism, adaptability, and fault tolerance when compared to a traditional single robot,” said Rubenstein.

The algorithm is based on technology similar to GPS, where each robot is aware of where it sits on the grid. Before making a decision about where to move, each robot uses sensors to communicate with its neighbors, determining whether or not nearby spaces within the grid are vacant or occupied. “The robots refuse to move to a spot until that spot is free and until they know that no other robots are moving to that same spot,” Rubenstein said. “They are careful and reserve a space ahead of time … each robot can only sense three or four of its closest neighbors, they can’t see across the whole swarm, which makes it easier to scale the system. The robots interact locally to make decisions without global information.”

Search and Rescue

In my opinion, one of the most important applications for swarms is when an area is deemed too dangerous for human exploration teams, search-and-rescue drones can survey a scene with camera-equipped unmanned fliers.

Teams from TU Delft, the University of Liverpool, and Radboud University of Nijmegen, created software that allows a swarm of tiny flying robots to autonomously explore an unknown environment. The robots navigate their environment by following the walls. With four laser range sensors in all four directions of the aircraft's horizontal plane, the drones determine a wall angle by evaluating whether the side range sensors are triggered in combination with the front one.

The tiny aircraft avoid collision and maximize their search efficiency by "talking" with each other via wireless communication chips. The signal strength between the chips is shown much like how the bars on our phones decrease when you move away from your Wi-Fi router at home. By sensing proximity, the drones keep their distance from each other, avoiding crashes.

Nature — The Great Teacher

Most of the researchers investigating swarms say they were inspired by nature. An article  by Chantal Nguyen, Postdoctoral Associate at the BioFrontiers Institute, University of Colorado Boulder, delves into natural swarms.

“Spontaneous pattern formation is a neat example of self-organization in nature. Self-organization refers to when initially disordered systems, such as a jungle of plants or a swarm of bees, achieve order without anything controlling them. Order emerges from the interactions between individual members of the system and their interactions with the environment. Somewhat counterintuitively, noise — randomness — facilitates self-organization.”

Nguyen uses the example of a colony of ants converging on a food source. “Ants secrete pheromones behind them as they crawl and other ants follow the pheromone trails, and further reinforce the trail they took by secreting their own pheromones.”

Interestingly, “If a few ants were to randomly deviate from the trail, they might stumble onto a shorter path and create a new trail. So, this randomness injects a spontaneous change into the ants’ system that allows them to explore alternative scenarios. Eventually, more ants would follow the new trail, and soon the shorter path would prevail. This randomness helps the ants adapt to changes in the environment, as a few ants spontaneously seek out more direct ways to their food source.”

Cooperation among animal organisms to form swarms of insects, fish, and birds has often been cited — but less so, spontaneous coordination among plants. Nguyen cites a 2017 study of sunflowers that found when they were grown in a dense row, they “naturally formed a near-perfect zigzag pattern, with each plant leaning away from the row in alternating directions. This pattern allowed the plants to avoid shade from their neighbors and maximize their exposure to sunlight.”

She and her team of researchers investigated the underlying mechanisms of this phenomenon. They discovered that some plants gradually move over time, a process termed circumnutation, originally observed by Charles Darwin. Since plants tend to face toward maximum sources of light, their random circumnutations eventually led to an organization that maximized the light for each plant — the zigzag.

Final Thoughts

What this all adds up to for me is that there is a lot to learn about how individual units can combine to form collective patterns but still retain individual control.

I’ve mentioned a few of the practical applications for researching swarm behavior, but the most important is the emerging world of autonomous vehicles — literally a matter of life and death.

---