In the near future, researchers hope to produce swarms of autonomous tiny robotic sea turtles and fishes, for example, to perform hazardous missions such as detecting nuclear waste underwater or other tasks too dangerous for humans.
Among underwater robotics, turtle robots are among the most maneuverable. A team of engineers have built a prototype turtle robot to perform set tasks, while also being able to react to exigencies and obstacles.
Associate Professor S K Panda of the National University of Singapore leads the electrical and computer engineering department team, which is involved in the research and development of biomimetic solutions in ocean engineering.
They’re looking to nature for solutions to technical challenges with robots that mimic natural systems. The team is currently putting the final touches on a robotic sea turtle that could move about underwater, including diving to deeper depths vertically, like a real turtle, with just its front and hind limb movement.
“Our turtle robot does not use a ballast system which is commonly used in underwater robots for diving or sinking functions,” says Panda.
“Without this ballast system, it is much smaller and lighter, enabling it to carry bigger payloads so that it can perform more complicated tasks such as surveillance, water quality monitoring in Singapore reservoir or energy harvesting for long endurance.
“Being able to do a dynamic dive or sinking vertically means that it can also enter vertical tunnels or pipes in the seabed with very small diameters.”
Being smaller and lighter would also enhance its energy efficiency. The new turtle robot is also able to self-charge, further reducing the need for it to return to base station for recharging. It is agile and able to turn sharp corners with small radius, without losing speed.
“We can have a swarm of tiny turtles which communicate with each other and act collaboratively to perform their duties. With improved maneuverability they can go to tiny and narrow places like crevices where bigger vessels are unable to do so.”
Fish prototypes, too
Abhra Roy Chowdhury, who has been working toward lifelike underwater robots for the last three years for his PhD, says the team has designed and developed four other underwater prototypes—a spherical robot that mimics a puffer fish in structure but uses a jet propulsion technique similar to jellyfishes and squids; and three different robotic fish. These robots are scalable, modular, and have stealth features.
“If need be, we can actually combine all their merits in a single robot,” adds Chowdhury.
Chowdhury first developed a lifelike fish robot about three years ago—after spending a considerable amount of time studying the maneuverable and energy efficient movements of real fish.
“For example, many fishes using body-caudal fin locomotion, bend their bodies into a backward-moving propulsive wave that extends to caudal fin while fishes using median-paired fin locomotion use their other fins like dorsal and anal to propel themselves,” he explains.
He had studied the yellow-fin tuna and the freshwater largemouth bass specifically, as they have the most common fish body types as well as swimming patterns. Both are found to be efficient swimmers in sea-environment. He has further developed a novel bio-inspired dynamics and behavior based control architecture for these biomimetic platforms.
Another member of the team, research engineer Bhuneshwar Prasad, has also developed a spherical robot. This robot can be used for oceanic surveys, inspections of pipe and cable, inspection of a ship hull or a propeller’s shaft, for example. The spherical robot is vision-based and uses a “visual servoing” system comprising of an on-board bottom facing camera module that is color-coded to extract the position information and then guide the robot.
“The spherical underwater robot, using a self-ballast system, is able to dock on soft ground to harvest energy from underwater currents. Once on the seabed, the robot can be placed in the sleep mode, with only monitoring sensors awake, to harvest power from the underwater current through dynamo-based rotor blades,” he explains.
“We expect to invent robots capable of performing collaborative intervention missions three to five years down the road. What we plan to do in the near future is to develop robot fish with muscles, which can undulate the way real fish do.”
“For this, we need to develop special actuators. We also aim to develop central pattern generators which will enable the fish to respond to external stimuli so that it can make crucial decisions to complete a critical mission,” says Panda.
Source: National University of Singapore