Looking to the natural world for engineering inspiration is an idea at least as old as Leonardo da Vinci. Copying nature directly, though, has often proved hard. For example, birds flap their wings to achieve both lift and propulsion, but flying machines that imitate this action have tended not to do well. Human engineering has found it easier to create aircraft by giving them fixed, rigid wings and propelling them with motors.
Air is not the only medium through which animals move by flapping, however. Many creatures flap wings, or wing-like structures, to “fly” through water. That is something human engineers can aspire to imitate because the buoyancy of water provides free lift and its density makes propulsion easier. And, as they write in this week’s Science, a group led by Kit Parker of Harvard University have done just that. They have built a robotic stingray (pictured above) that imitates the motion of its biological counterpart. Moreover, it does so not with the electric circuits and servomotors of conventional robots, but with muscle cells engineered to mimic the elegant undulations of a living stingray.
The ray itself is a so-called soft robot, a type of ’bot that has gained prominence of late. Soft robots, which are made of materials such as latex and silicone, are able to squeeze through tight openings, handle fragile objects and interact with humans far more safely than can their rigid metal and plastic counterparts. Yet most soft robots are propelled by pneumatic pressure or cables that are, in turn, driven by bulky, rigid motors. Having real muscles instead of these ersatz ones would make much more sense for a soft robot, since muscles, too, are soft, and are powered by glucose, not motors.
Dr Parker and his team chose rat muscles for their rays. They grew rat-muscle cells in culture and then “printed” them onto sheets of elastomer that were to act as the surfaces of a robo-ray’s wings. Muscle cells work by contracting, which is why muscles often operate in pairs (like the biceps and triceps of the arm), the elements of which pull in opposite directions. To simplify things, though, the team used only one layer of muscle, to pull in one direction. The opposite pull was supplied by the relaxation of a skeleton of gold that had been put under tension by the muscle cells’ initial compression. The result, when cut into an appropriate shape, was a plausible facsimile of a ray.
To co-ordinate the muscle cells’ contraction in a way that would propel the ray forwards, Dr Parker printed them in serpentine patterns. When one cell was activated, it released calcium ions which acted (as happens in nature) as a signal to the next cell in the sequence to contract. Thus, like a cascade of dominoes, waves of muscular undulation passed from one end of the ray to the other.
Perhaps the cleverest component of the bionic ray, though, was its control mechanism. The muscle cells Dr Parker chose were genetically engineered so that light activated their contraction. Flashing a beam at the front of the ray’s fins caused an undulation to start propagating. Each new flash triggered a new undulation, so making the ray move forward in a straight line meant flashing at each fin simultaneously. Increasing the rate of flashing on one side of the robot but not the other caused that side to flap faster, turning the machine away from the hyperactive fin. This let the team steer the device. Thus controlled, the 16mm-long ray was able to travel at 90mm a minute, and to complete a 250mm slalom course at that speed without touching any of the obstacles. Moreover, it was able to do so on six consecutive days, retaining 80% of its original speed on the final day of the trial.
What use might be made of a more sophisticated version of such a robot remains to be seen. Dr Parker’s design requires the fluid the ’bot swims through to contain the glucose that power its muscles, reducing the ’bot’s deployability. But a future version might be fitted with a glucose reservoir and a cardiovascular system to circulate that glucose. This would let it go anywhere which had enough oxygen in the water for the muscle cells to respire and would make it resemble a real animal even more closely than it does already.
Air is not the only medium through which animals move by flapping, however. Many creatures flap wings, or wing-like structures, to “fly” through water. That is something human engineers can aspire to imitate because the buoyancy of water provides free lift and its density makes propulsion easier. And, as they write in this week’s Science, a group led by Kit Parker of Harvard University have done just that. They have built a robotic stingray (pictured above) that imitates the motion of its biological counterpart. Moreover, it does so not with the electric circuits and servomotors of conventional robots, but with muscle cells engineered to mimic the elegant undulations of a living stingray.
The ray itself is a so-called soft robot, a type of ’bot that has gained prominence of late. Soft robots, which are made of materials such as latex and silicone, are able to squeeze through tight openings, handle fragile objects and interact with humans far more safely than can their rigid metal and plastic counterparts. Yet most soft robots are propelled by pneumatic pressure or cables that are, in turn, driven by bulky, rigid motors. Having real muscles instead of these ersatz ones would make much more sense for a soft robot, since muscles, too, are soft, and are powered by glucose, not motors.
Dr Parker and his team chose rat muscles for their rays. They grew rat-muscle cells in culture and then “printed” them onto sheets of elastomer that were to act as the surfaces of a robo-ray’s wings. Muscle cells work by contracting, which is why muscles often operate in pairs (like the biceps and triceps of the arm), the elements of which pull in opposite directions. To simplify things, though, the team used only one layer of muscle, to pull in one direction. The opposite pull was supplied by the relaxation of a skeleton of gold that had been put under tension by the muscle cells’ initial compression. The result, when cut into an appropriate shape, was a plausible facsimile of a ray.
To co-ordinate the muscle cells’ contraction in a way that would propel the ray forwards, Dr Parker printed them in serpentine patterns. When one cell was activated, it released calcium ions which acted (as happens in nature) as a signal to the next cell in the sequence to contract. Thus, like a cascade of dominoes, waves of muscular undulation passed from one end of the ray to the other.
Perhaps the cleverest component of the bionic ray, though, was its control mechanism. The muscle cells Dr Parker chose were genetically engineered so that light activated their contraction. Flashing a beam at the front of the ray’s fins caused an undulation to start propagating. Each new flash triggered a new undulation, so making the ray move forward in a straight line meant flashing at each fin simultaneously. Increasing the rate of flashing on one side of the robot but not the other caused that side to flap faster, turning the machine away from the hyperactive fin. This let the team steer the device. Thus controlled, the 16mm-long ray was able to travel at 90mm a minute, and to complete a 250mm slalom course at that speed without touching any of the obstacles. Moreover, it was able to do so on six consecutive days, retaining 80% of its original speed on the final day of the trial.
What use might be made of a more sophisticated version of such a robot remains to be seen. Dr Parker’s design requires the fluid the ’bot swims through to contain the glucose that power its muscles, reducing the ’bot’s deployability. But a future version might be fitted with a glucose reservoir and a cardiovascular system to circulate that glucose. This would let it go anywhere which had enough oxygen in the water for the muscle cells to respire and would make it resemble a real animal even more closely than it does already.