A team of scientists and engineers at Polytechnic Institute of New York University (NYU-Poly) reports significant progress in devising methods for leading live fish away from oil spills and other aquatic dangers using a species-specific robotic fish. The researchers say they now also see the possibility of extending their results into the world of biomedical research.
The team's latest research centered the visual response of zebrafish to a robotic fish that was designed and animated to attract their attention. The results were reported today in IPO Publishing’s academic journal Bioinspiration and Biomimetics. This study followed one on another species, golden shiners that investigated fish response to hydrodynamic cues generated by a robot. In keeping with the differences between the two species, the new study's findings also were quite distinctive, said Maurizio Porfiri, associate professor in the NYU-Poly Department of Mechanical and Aerospace Engineering.
The interdisciplinary team also includes NYU-Poly research scholar Giovanni Polverino, doctoral candidates Nicole Abaid and Vladislav Kopman and scientist Dr. Simone Macrì of the Istituto Superiore di Sanità in Rome.
In the prior study, golden shiners swam behind the robotic fish according to the wake created by the robot's undulating tail; the hydrodynamics of the wake offered the fish an energy-saving advantage, Porfiri explained.
In the current study, which was conducted in an enclosed tank, the robotic fish remained in place and was separated from the zebrafish by transparent barriers. Although the robot moved its tail, the zebrafish swam in a placid environment and experienced no such swimming advantage from the robot. Yet the zebrafish did congregate at the barrier near the robot, proving that the appearance of the robotic fish was itself a factor in the attraction, Porfiri said.
The team fashioned their robotic fish to look like a fertile female zebrafish, which is attractive to both males and females in the species, and they painted it with characteristic stripes and yellow accents. However, the robot's size was much larger than that of a live zebrafish: six inches long versus one inch, respectively.
There were four important findings, Porfiri says.
- When the robotic fish with flapping tail was adjacent to an empty control section, both individuals and small shoals of zebrafish preferred to spend time in the vicinity of the robotic fish. In other words, fish were attracted by the robot.
- When the robotic fish was adjacent to a replica whose tail was still, the zebrafish moved to the side of the tank near the moving robot, indicating that motility is attractive.
- When the robotic fish with flapping tail was adjacent to an empty control section and the tank was dark, the zebrafish moved toward the empty compartment, showing that they feared the noise generated by the robot's motor, but also that the motility of the tail wasn’t attractive if it was not seen. Or, stated differently, when visual and auditory stimuli are together, "the visual stimulus is so strong that it can overcome what they hear," Porfiri said.
- When a single zebrafish was placed in the control section, zebrafish flocked to it rather than to the robotic fish, demonstrating that they recognized the difference between the two and preferred their peer.
Thus, for future studies, Porfiri said, one goal will be to produce robotic fish that flap their tails silently.
Other directions for the team's future research include building robotic fish that are equipped with cameras and imbued with artificial intelligence, so they can autonomously change their behavior in reaction to the behavior of the fish being studied; and placing different species of fish together in an attempt to attract one species and repel the other at the same time. Right now, "we don't have a systematic way to change the behavior of the robot as a function of what the animal is doing," Porfiri said. "We want to use it to understand different hypotheses of leadership and communication among animals."
In addition, Porfiri said, the team wants to explore how it can apply its robotic zebrafish in the biomedicine industry, which already uses live zebrafish as an aquatic analog to mice for testing drugs under development. For example, he postulates, the NYU-Poly robotic zebrafish could provide a reliable stimulus when studying the effect of a drug on social behavior.
Partial funding for this latest research was granted by the National Science Foundation. Other financial support was provided by the Honors Center of Italian Universities (H2CU) that partially supported NYU-Poly's Polverino. The journal article is entitled, "Zebrafish Response to Robotic Fish: Preference Experiments on Isolated Individuals and Small Shoals."