Robotic fish learns to match its swimming speed to the current
Fish have a sensory system known as the lateral line, which allows them to detect movements, vibrations and pressure gradients in the water. Scientists have now given a robotic fish its own version of that system, letting it determine the best swimming speed.
The study involved researchers from the Max Planck Institute for Intelligent Systems (Germany), Seoul National University and Harvard University. They created a soft-bodied fish-inspired robot, which was able to swim in place against a water current passing through a tank.
Its undulating swimming motion was made possible thanks to a series of linked silicone chambers, located along either side of its body. Air was alternately pumped into the chambers on one side and out of those on the other – this caused the inflated side to expand and curve outwards, while the deflated side curled inwards.
The robot's lateral line system consisted of two liquid-metal-filled silicone microchannels, running the length of each side. As each of those channels stretched while that side of the body curved, the electrical resistance of the liquid metal within increased. Therefore, by monitoring the changes in resistance, it was possible to determine how much a given amount of air pressure caused the robot's body to undulate.
The scientists proceeded to set up a self-learning loop, in which a computer connected to the robot measured the changing water current velocity, then automatically adjusted the air pressure in response to that information. Doing so allowed the robot to continuously maintain a swimming speed which matched that of the current. In a natural environment such as a river, this would keep the robot from being swept downstream when not proceeding forward.
"This robot will allow us to test and refine hypotheses regarding the neuromechanics of swimming animals as well as help us improve future underwater robots," says Max Planck's Dr. Ardian Jusufi. "In addition to characterizing the soft strain sensor under submerged dynamic conditions for the first time, we also developed a simple and flexible data-driven modelling approach in order to design our swimming feedback controller."
A paper on the research was recently published in the journal Advanced Intelligent Systems. The robot can be seen in action, in the video below.