Researchers at Harvard’s Vice Institute for Biological Inspired Engineering have created a set of fish-shaped underwater robots that can collaborate to perform tasks, to autonomously navigate each other and find each other. Or may choose only school together.
As aerial drones are proving themselves useful in industry after industry, underwater drones can revolutionize ecology, shipping and other areas, where a continuous underwater presence is desirable but difficult.
The last few years have seen interesting new autonomous underwater vehicles, or AUVs, but the most common type is a torpedo – efficient for open-water cruises, but working through a coral reef or marina’s nook and crane Not to do.
For that purpose, it is practical to see what nature has seen fit to create itself, and the Wyss Institute has characterized doing so and building robots and machinery in imitation of the natural world.
In this case Florian Berlinger, Melvin Gauci, and Radhika Nagpa, who are all co-authors on a new paper published in Science Robotics, decided not only to mimic the shape of a fish, but the way it was from other peers. Interacts with.
Inspired by the sight of schoolfishes during scuba diving, Nagpa pursued the question: “How do we create artificial agents that can demonstrate such collective coherence where an entire collective feels as if it were a single agent is?”
His answer, Bluesworm, is a collection of small “bluebots”, 3D-printed in fish shape, with wings instead of propellers and cameras for the eye. Although neither you nor I are likely to mistake these for real fish, they are less scary than an object for a normal fish to see a six-foot metal tube that is placed on a propeller Spins loudly along. Blueboats mimic nature’s bioscience innovation, lighting up with LEDs the way some fish and insects signal to others. LED Pulses change and adjust based on the knowledge of each bot’s position and its neighbors.
Using the cameras’ simple senses and so forth, an early swimming motions, and LEDs, Bluesworm automatically organizes itself into group swimming behavior, establishing a simple “milling” pattern that allows new bots Adjusts when dropped from any. Angle.
Robots can also work together on simple tasks, such as searching for something. If the group is given the task of finding a red LED in the tank, they can watch each one independently, but when one of them finds it, it makes its own LED glow to alert and summon the others. Replaces
It is not difficult to imagine the use for this technique. These robots can monitor their health or detect their camera eyes, without endangering marine life. Or they can diving downwards and ships inspecting ships can be seen more efficiently than a single craft. Perhaps they can also be useful in search and rescue.
This research advances our understanding of how and why animals swim together in the first place.
With this research, we can not only build more advanced robot collectives, but also learn about collective intelligence in nature. Fish should follow even simpler behavior patterns when swimming in schools than our robots. Berlinger said that this simplicity is so beautiful that it is difficult to find. “Other researchers have already reached out to me to use my blueboats because the fish uses fish for biological studies on swimming and schooling. The fact that they welcome bluebots among their laboratory fishes makes me very happy. “