These robotic fish will allow monitoring of underwater environments and pace the way for future advances in autonomous vehicles.
“We’re looking at how to coordinate the robots using only on-board vision sensing and minimal communication,” Morgansen said. “In the ocean, the ability to transmit signals through the water is extremely limited, and lack of light makes full computer vision difficult. One of the things we’re interested in is what we can do with much more rudimentary sensing such as using low-resolution imagery.”
In addition to being able to improve aquatic monitoring systems, the robotic fish have provided a prototype to study various aspects of autonomous vehicles. Some of these control systems implemented in the fish include collision and obstacle avoidance. Nathan Powel, a research assistant in the Nonlinear Dynamics and Control Lab, described the potential to apply such mechanisms to a wide range of vehicles including fixed wing aircraft and helicopters.
“We are focusing on developing control algorithms to make autonomous vehicles more capable,” Powel said. “The fish gave us a place to test those systems.”