Robotic Swarm Learns to Swim Like a School of Fish

Certain fish swarm in schools and swim with synchronized behaviors to help find food, migrate, and evade predators. There are no leaders or a team of fish that coordinate that schooling behavior, nor is there any communication between them on what to do while in those schools. Instead, these fish rely on ‘implicit coordination,’ or the ability to make decisions based on what they see others doing in a said swarm. Think of it as self-autonomous organization and coordination, otherwise known as ‘just following the crowd.’

Researchers from Harvard’s John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Wyss Institute for Biologically Inspired Engineering have designed robotic fish that can synchronize their movements and form schools based on their biological counterparts. The robotic fish are known as ‘Bluebots’ and are outfitted with a pair of cameras and several blue LEDs. The robots use the fisheye cameras to sense the LED light emitted by neighboring Bluebots, then use custom algorithms to gauge their distance, direction, and heading in a 3D space (in this case, water).

“Each Bluebot implicitly reacts to its neighbors’ positions,” states Ph.D. candidate Florian Berlinger of the Wyss Institute. “So, if we want the robots to aggregate, then each Bluebot will calculate the position of each of its neighbors and move towards the center. If we want the robots to disperse, the Bluebots do the opposite. If we want them to swim as a school in a circle, they are programmed to follow lights directly in front of them in a clockwise direction.”

The researchers simulated a search mission for the robotic fish using a red light positioned in a water tank, which is used for a focal point, or visual target. They then engaged the disperse algorithm, enabling the fish to spread out in the tank until one of the Bluebots gets close enough to the red light to detect it. Once detected, the initiator robot begins to blink its blue LEDs, which then triggers the aggregation algorithm in the rest of the robots, who then converge on the initiator.

The researcher’s work is a milestone in collective underwater robotic behavior, which will help develop small underwater swarms that could be tasked with environmental monitoring or search missions in fragile ecosystems, such as coral reefs.