Fish-inspired robots coordinate uncontrolled external movements

Fish ladders display complex, synchronized behaviors that help them find food, migrate and escape from predators. No single fish or team of fish coordinates these movements and fish do not communicate with each other about what should be done next. Instead, this common behavior stems from understandable coordination – individual fish making decisions based on what they see their neighbors doing.

This kind of self-organizing and self-regulating, self-organizing coordination has attracted a great deal of interest of scientists, especially in the field of robots.

Now, a team of researchers at Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Wyss Institute for Biologically Stimulated Engineering have inspired robots with fish that can synchronize their movements like a real school of fish, with no outside control. This is the first time researchers have demonstrated 3-D complex behavior with apparent coordination in underwater robots.

“Robots are often used in areas that are accessible or dangerous to humans, areas where human intervention would not be possible,” said Florian Berlinger, Ph.D. Candidate at SEAS and Wyss and first author of the paper. “In these situations, you have the advantage of having a highly automated, self-sufficient robot swarm. Using comprehensible rules and 3-D visual understanding, we were able to create a system has a high degree of underwater independence and flexibility, where things like GPS and WiFi are not accessible. “

The research is published in Robotics Science.

The fish-inspired robotic sword, called Blueswarm, was created in the laboratory of Radhika Nagpal, Fred Kavli Professor of Computer Science at SEAS and Associate Faculty Member at the Wyss Institute. Nagpal Laboratory is a pioneer in self-organizing systems, from their 1,000 Kilobot robot swim to the termite-inspired robotic construction team.

However, most of the previous robotic swarms operated in a two-dimensional space. Three-dimensional spaces, such as air and water, pose significant challenges to perception and movement.

To overcome these challenges, the researchers developed a vision-based coordination system in their fish robots based on blue LED lights. Each underwater robot, called the Bluebot, is equipped with two cameras and three LED lights. The cameras on board a lensed fish detects neighboring Bluebots LEDs and use a custom algorithm to determine their speed, direction and head. Based on simple production and LED light detection, the researchers showed that the Blueswarm could exhibit complex self-organizing behaviors, including assembly, scattering, and circle formation.

“Every Bluebot suddenly reacts to the roles of its neighbors,” Berlinger said. “So if we want the robots to go together, then each Bluebot measures the position of each of its neighbors and moves toward the center. If we for the robots to spread, the Bluebots will do the opposite. If we want them to swim like a ladder in a circle, they are programmed to follow lights directly in front of them in a clockwise direction. “

The researchers also considered a simple survey mission with a red light in the tank. Using the scattering algorithm, the Bluebots scatter out across the tank until one comes close enough to the light source to detect it. As soon as the robot finds the light, its LEDs will start flashing, which will trigger the collection algorithm in the rest of the school. From there, all the Bluebots go around the signal robot.

“Our results with Blueswarm represent an important milestone in the study of self-organized underwater collection behavior,” Nagpal said. “Views from this research will help us develop small future underwater flocks that can study and study the environment in high-visibility but fragile environments such as coral reefs. This research will also be a way for schools better understand fish, by reproducing their behavior synthetically. ”