Few things make the career jump from regional cuisine delicacy to robotics actuator. Yet, in research vaguely reminiscent of RoboCop, scientists have used skeletal muscle from a literal frog leg as an actuator to control a tiny robotic manta ray.
Researchers at the Shenyang Institute of Automation (SIA) of the Chinese Academy of Sciences have built a wireless, light-controlled manta ray robot whose fins are driven by actual biological muscle. Taken from the gracilis muscle in the leg of a bullfrog, the tissue produced a stable contractile force of 6.5 newtons (N), peaked at 9.4 N under more extreme stimulation, remained electrically responsive for as long as 11 days, and could reliably drive the robot over a seven-day period.
The research falls under biosyncretic robotics, an emerging field of biohybrid robotics that incorporates living biological material into machines. The biological components perform functions typically handled by electromechanical components, replacing complex, bulky systems. In this case, biological muscles are serving as actuators, replacing electric motors, pneumatics, and hydraulics.
Muscles are particularly suited for biosyncretic robots, as nature has already spent hundreds of millions of years developing a compact, compliant, and highly efficient actuator. A skeletal muscle can contract strongly and rapidly, over and over again, while simultaneously functioning as its own extremely sophisticated mechanical transmission.
Researchers have consequently built various muscle-powered robots, using synthesized muscle tissue. The SIA team itself produced an earlier manta-inspired swimmer in 2022 using lab-grown skeletal muscle tissue. However, a notable problem remained. Engineered muscle tissues reconstructed from cultured cells are comparatively weak. According to the researchers, low output force is one of the factors that has limited the speed and maneuverability of existing skeletal-muscle-powered robots.
So this time around, they skipped the muscle-creation process and borrowed a finished version that evolution had perfected. The team isolated the gracilis, a long skeletal muscle found in the leg, from a bullfrog. Unlike reconstructed muscle bundles, the native tissue retains its naturally organized arrangement of muscle fibers. The researchers say this ordered architecture provides considerably better contractile performance.
They then tested different electrical signals to activate it. At an optimized stimulation of 1 Hz, 5 V, and 10 milliseconds per pulse, the muscle produced about 6.5 N of contractile force and shortened by around 25%. Under more extreme conditions, peak force reached 9.4 N.
“For the first time, we have coupled native isolated skeletal muscle with a wireless optoelectronic neural-stimulation system,” says SIA researcher Dr. Chuang Zhang. “Our work validates native muscle tissue as a high-performance biological actuator."
Now, having a strong biological actuator only solves half of the problem. You still need to power it and tell it when to move. In animals, adenosine triphosphate (ATP) inside the muscle supplies most of the mechanical energy, while an electrical impulse travels along a motor neuron and triggers a chain of events that causes skeletal muscle fibers to contract.
Biohybrid robots replicate that process with electrodes that deliver electrical pulses to the muscles. The obvious problem for a mobile robot is where those electrical pulses come from. Conventional setups commonly relied on external electrodes and wired stimulators that tethered the robots to external systems. The SIA team's earlier manta-ray robot swam inside an electrode-equipped environment, with eight platinum electrodes arranged around it to generate the electric fields that stimulated its cultured muscle. It worked, but the swimming environment itself effectively formed part of the robot's control system.
The new manta ray robot solves that problem with an ingenious approach.
Tiny gallium arsenide (GaAs) photovoltaic modules sit on the robot's dorsal (upper) surface. An operator shines an 808-nm near-infrared laser onto them. The cells convert that incoming light into electricity, and the resulting electrical signal travels to nerves on the surface of the frog muscle, causing it to contract. The left and right sides can also be addressed separately. By controlling when and where the NIR light hits the two photovoltaic sections, the researchers can control the contraction sequence of the muscles on either side, achieving untethered control.
Real manta rays propel themselves by rhythmically flapping and undulating their enormous pectoral fins. It's a well-suited biological template for this kind of actuator, as contracting muscles can be mechanically coupled to flexible fins, turning repeated shortening and relaxation into a flapping swimming motion.
The researchers similarly arranged the isolated muscle actuators to drive the robot's pectoral fins. As the electrical stimulation triggers contraction, the muscle pulls on the structure, causing the fins to flap and generate thrust. Controlling the two sides together drives the robot forward, while changing their activation patterns creates asymmetric thrust for steering.
The robot achieved an average straight-line speed of 0.54 body lengths per second, equivalent to roughly 2.7 cm (1.1 in) per second, and briefly reached two body lengths per second. The researchers say the latter is the fastest relative forward speed yet reported for a skeletal-muscle-driven biosyncretic robot.
It could turn within just one-eighth of its own body length, reached a maximum angular turning speed of 21 degrees per second, and completed an entire circular trajectory in about 17 seconds. Selectively stimulating its left and right sides allowed straight swimming, left and right turns, circles and U-turns. The frog-ray could also cruise at the water surface, swim underwater while carrying a payload of up to 5 g (0.18 oz) and, rather unexpectedly for a manta, produce slow curved movements across a rigid surface.
Now, the relevance of the work extends beyond the selected anatomy. Neither of the innovations, the use of natural tissue and the method of powering the actuator, is inherently restricted to manta-shaped swimmers. Native isolated skeletal muscles could conceivably drive other robot architectures, while the light-to-electricity control technique provides a way to command biological actuators without permanently wiring the robot to an external stimulator.
For this particular machine, the researchers see possible applications in shallow-water environmental monitoring and low-disturbance observation of aquatic organisms; situations where a small, fin-propelled machine could move through an environment without the propellers and rigid mechanisms found on conventional underwater robots.
The study also produced further benefits. The team says the electrical-stimulation framework developed to keep isolated muscle contracting effectively could serve as an experimental reference for in-vitro muscle rehabilitation following nerve injury and for dynamic tissue-engineering culture systems
For now, an obvious limitation is the lifespan of the biological tissue. The integrated muscle could operate the robot reliably for about seven days and remained electrically responsive for up to 11 days. Furthermore, the current machine also depends on externally supplied near-infrared light for its wireless electrical stimulation. While this gets rid of a physical tether, it still leaves the robot dependent on an external energy/control source.
This brings us to the team's next steps. The researchers plan to develop better in-vitro culture strategies to extend muscle life, as well as neuron-like flexible thin-film electrodes that can distribute electrical stimulation more uniformly across the tissue. They also want to combine the photovoltaic system with onboard energy storage, allowing the robot to store harvested light energy and use it later rather than keeping its movements directly tied to incoming illumination.
The study was published in the journal Advanced Functional Materials.
Source: Chinese Academy of Sciences
