Simple Instructions Make a Robot Swarm
It’s often said that the whole is greater than the sum of its parts.
Certainly, in nature, animals that move in herds, flocks, or swarms gain competitive advantages in terms of access to mates, cover from predators, and more time to forage for food. Over millennia, different species have evolved unique movements in order to “move as one” and reap the rewards of coming together. Think of the spiraling swims of schools of fish or the V-formation of migrating geese.
Now, researchers from the University of Texas have leveraged Mother Nature’s innate collective behavior by training micro-robots, or particles powered by optothermal fields, to form “smart swarms,” where each micro-robot adapts its motion based on the environment in order to form a collective structure.
Yuebing Zheng said that if robots can become better at collective motion, “they can become better at navigation and achieve better function collectively.”
“If we look to nature and these kinds of swarms of bees and schools of fish, they aren’t just moving randomly,” said Yuebing Zheng, associate professor in the Walker Department of Mechanical Engineering and Texas Materials Institute at UT Austin. “They actually form very interesting patterns based on what their neighbors are doing—and, from a functional point of view, their being able to understand this pattern is very important for their survival. And it’s the same for these synthetic microrobots. If they can adopt this kind of collective motion, they can become better at navigation and achieve better function collectively.”
To program such collective behavior, Zheng and colleagues looked to nature and how groups of animals learn to move as a group and then applied it to an optical feedback-control system.
“Fish, for some reason, can form this very organized swimming pattern, even though there is no one leader saying, ‘Hey! Everyone follow me!’” Zheng explained. “The reason that fish in a school achieve collective motion is each fish in the school follows the same simple rules. Every fish follows its neighbor. When they do that, they can achieve the right pattern, even when the environment changes and they have to adapt their response.”
Discover the Benefits of ASME Membership
The research team placed the micro-robots in a static liquid solution. Each robot’s motion was driven by an individual laser beam as part of an optical feedback-control system. As each robot moved within the swarm and interacted with its neighbors, it was governed by the same rules as the fish—follow thy neighbor. In doing so, it adapted its motion to changes in the local surroundings. A response that Zheng and colleagues referred to as “adaptive time delay.”
“The essential strategy is adaptive time delay,” Zheng said. “You control the time of one fish to receive the information of the neighboring fish, calculate the time difference, and the adjustment based on that time difference is universal.”
Zheng said this proof of concept worked extremely well in a well-controlled laboratory setting—and he and his colleagues plan to test it in “noisier” settings in the future.
“This was a relatively ideal experiment. We didn’t even have water flow,” he said. “It was a static setting, and we simply changed the time response of individual robots by controlling the external light signal. In reality, however, this type of collective motion happens in the ocean or in the air where the environment is very dynamic. So, we need to move this system into a flow environment to see how it will affect collective motion and how the adaptive time delay could be implemented to overcome other disturbances so groups of robots can still maintain a robust collective motion.”
Zheng said he sees a future where such nanorobots could be used to clean pollutants from water sources, using chemical reactions driven by sunlight, or perhaps for medical applications to detect tumors in the body or deliver targeted drug therapies. He also said it is possible they could be used for autonomous vehicles to help avoid accidents or traffic jams.
More Like This: A School of Fish is Quieter than Just One
These types of applications may seem “futuristic,” but Zheng said that if we can figure out the control mechanisms, they are indeed possible. It will just take time and hard work.
“There are many future scenarios, but you need to make sure these collective structures are both robust and responsive,” he said. “You never know what will happen when you have a group of robots moving together and an unexpected obstacle comes. You want to be sure they know what to do and can avoid any problems. So, we will continue to be inspired by nature, appreciate what nature can do, and learn what we can. We know we must keep a curiosity about how nature does these incredible things so we can apply what we have learned to solve actual problems for the future.”
Kayt Sukel is a business and technology writer in Houston.
Certainly, in nature, animals that move in herds, flocks, or swarms gain competitive advantages in terms of access to mates, cover from predators, and more time to forage for food. Over millennia, different species have evolved unique movements in order to “move as one” and reap the rewards of coming together. Think of the spiraling swims of schools of fish or the V-formation of migrating geese.
Now, researchers from the University of Texas have leveraged Mother Nature’s innate collective behavior by training micro-robots, or particles powered by optothermal fields, to form “smart swarms,” where each micro-robot adapts its motion based on the environment in order to form a collective structure.
Yuebing Zheng said that if robots can become better at collective motion, “they can become better at navigation and achieve better function collectively.”
To program such collective behavior, Zheng and colleagues looked to nature and how groups of animals learn to move as a group and then applied it to an optical feedback-control system.
“Fish, for some reason, can form this very organized swimming pattern, even though there is no one leader saying, ‘Hey! Everyone follow me!’” Zheng explained. “The reason that fish in a school achieve collective motion is each fish in the school follows the same simple rules. Every fish follows its neighbor. When they do that, they can achieve the right pattern, even when the environment changes and they have to adapt their response.”
Discover the Benefits of ASME Membership
The research team placed the micro-robots in a static liquid solution. Each robot’s motion was driven by an individual laser beam as part of an optical feedback-control system. As each robot moved within the swarm and interacted with its neighbors, it was governed by the same rules as the fish—follow thy neighbor. In doing so, it adapted its motion to changes in the local surroundings. A response that Zheng and colleagues referred to as “adaptive time delay.”
“The essential strategy is adaptive time delay,” Zheng said. “You control the time of one fish to receive the information of the neighboring fish, calculate the time difference, and the adjustment based on that time difference is universal.”
Zheng said this proof of concept worked extremely well in a well-controlled laboratory setting—and he and his colleagues plan to test it in “noisier” settings in the future.
“This was a relatively ideal experiment. We didn’t even have water flow,” he said. “It was a static setting, and we simply changed the time response of individual robots by controlling the external light signal. In reality, however, this type of collective motion happens in the ocean or in the air where the environment is very dynamic. So, we need to move this system into a flow environment to see how it will affect collective motion and how the adaptive time delay could be implemented to overcome other disturbances so groups of robots can still maintain a robust collective motion.”
Zheng said he sees a future where such nanorobots could be used to clean pollutants from water sources, using chemical reactions driven by sunlight, or perhaps for medical applications to detect tumors in the body or deliver targeted drug therapies. He also said it is possible they could be used for autonomous vehicles to help avoid accidents or traffic jams.
More Like This: A School of Fish is Quieter than Just One
These types of applications may seem “futuristic,” but Zheng said that if we can figure out the control mechanisms, they are indeed possible. It will just take time and hard work.
“There are many future scenarios, but you need to make sure these collective structures are both robust and responsive,” he said. “You never know what will happen when you have a group of robots moving together and an unexpected obstacle comes. You want to be sure they know what to do and can avoid any problems. So, we will continue to be inspired by nature, appreciate what nature can do, and learn what we can. We know we must keep a curiosity about how nature does these incredible things so we can apply what we have learned to solve actual problems for the future.”
Kayt Sukel is a business and technology writer in Houston.
The use of robotics in industrial settings continues to accelerate at a rapid pace. Explore this collection to learn more about automation, cobots, AI-enabled robots, customized robots, and drones.