Multifunctional fluidic units generate coordinated motion and sensing in soft robots by combining actuation, switching, and oscillation in one module, reducing reliance on electronic control.
(Nanowerk Spotlight) Soft robots offer the possibility of machines that move with gentle, adaptable motion instead of rigid, segmented steps. Early versions showed how silicone skins, inflatable chambers, and flexible joints could lift delicate objects or squeeze through narrow spaces. Yet these demonstrations also revealed a constraint.
Most soft robots still depended on clusters of electronic valves and processors mounted off the body, which pushed air or liquid into the robot through long bundles of tubing. That setup supplied power and control but limited independence and created a divide between the compliant body and the rigid hardware governing it.
Researchers looked for ways to embed more capability inside the soft structures themselves. Fluidic logic was an early step in that direction. These pressure driven circuits used channels and valves to time motions or trigger actions, reducing the need for electronics. They proved that a robot could use only air to coordinate behaviors, but they also required many distinct parts. One device sensed touch. Another oscillated to drive repeated motion. Others switched flows or produced bending. Each piece had to be fabricated and plumbed, often with custom tuning. As experiments grew more complex, the number of components and connections grew with them.
Advances in silicone casting, additive manufacturing, and small-scale pneumatic design created a new opportunity. These methods made it practical to build modules that changed function through connection rather than redesign. The idea echoed biological systems in which one structure often supports, senses, and moves at the same time. A single pneumatic building block with multiple behaviors could reduce the number of parts while expanding what soft robots could do.
This context frames the significance of the study published in Advanced Materials (“Multifunctional Fluidic Units for Emergent, Responsive Robotic Behaviors”). The paper introduces one compact fluidic unit that can behave as a valve, sensor, actuator, or self-oscillating limb. The hardware stays the same. Only the tubing connections change.
By combining several identical units on simple bodies, the researchers demonstrate hopping and crawling machines that coordinate motion and react to edges without electronic control. The work shows how one device can consolidate functions and produce system level behaviors through interaction, not computation.
Fluidic Unit Operation. A) The fluidic unit and its main components. B) Rendered images of the unit in its unactuated and actuated states, with an illustration of the sleeve cross-section showing flow pressure in both cases. C). The unit is configured as a sensor, actuator, or valve, with corresponding input/output connections. The middle panel shows the bending angle (𝜃) and the moment components, where M = F × D. D). The responsive self-oscillating actuator symbol and connection. Once pressurized, it oscillates at high frequency with small amplitudes; when an external moment is applied, the oscillations increase in amplitude and decrease in frequency. (Image: Reproduced from DOI:10.1002/adma.202510298, CC BY) (click on image to enlarge)
The unit itself contains a rigid housing, a flexible silicone sleeve, a 3D printed tube inside the sleeve, and a thin pneumatic pouch. The tube rotates inside the sleeve. Ports on the housing connect the internal tube and the pouch to outside lines. When the sleeve bends, the tube tip presses against the inner wall and blocks airflow. Because pressure can bend the sleeve and inflate the pouch, and those motions change the internal flow, the same mechanical structure supports several modes of operation.
In sensing mode, an air supply drives flow through the tube from an inlet to an outlet. The tube and sleeve extend outward. When an object touches them, they bend and the tube tip seals against the sleeve. The airflow drops sharply. That pressure change becomes a clear signal for downstream elements. This creates tactile sensing without electronics.
In actuation mode, only the pouch receives pressure. As it inflates, it produces a torque that rotates the sleeve and tube. This rotation can lift a weight or swing a segment of a limb. When pressure falls, the silicone elasticity resets the device. Motion arises from simple pressurization and relaxation.
In valving mode, the tube carries air between two ports while the pouch receives a control pressure. Rising control pressure inflates the pouch and bends the sleeve, which rotates the tube until it blocks the path. When control pressure falls, the sleeve straightens and the path reopens. The valve does not close and open at the same pressures. This difference is called hysteresis. It occurs because sealing the tube path creates a pressure difference between the inside and outside of the tube tip, which stabilizes the closed state. Hysteresis helps the valve switch cleanly and forms the basis of oscillation.
The study fixes one design for clarity in testing. It uses a DragonSkin 30 sleeve, a tube with an angled flat cut, and a two-segment polyethylene pouch. Even with a fixed design, the behavior is tunable through simple changes in connection and loading.
The most versatile configuration is a self-oscillating circuit that relies only on the unit itself. The inlet connects to a constant pressure source. The outlet supplies both the pouch and any external load. A controlled leak vents air slowly to the environment.
Together these form a relaxation oscillator, a cycle in which air accumulates, triggers a switch, then releases. During charging, air flows from inlet to outlet and inflates the pouch. At a threshold, the sleeve bends enough for the tube to block the path. Pressure at the outlet jumps to a higher level. During discharge, inflow stops and the pouch releases air through the leak until pressure falls below a second threshold. The tube then swings open and the cycle repeats.
This setup oscillates between about 75 kPa and 140 kPa with a frequency near 31 Hz. At higher pressure the pouch does not relax enough, and the motion stops. The researchers show that adding a counterweight to the tube tip introduces a restoring moment that opposes the pouch. With this simple change, the oscillator works at pressures up to at least 200 kPa and continues even at 400 kPa.
Frequency stays almost constant at a given weight. Heavier weights reduce frequency and increase swing amplitude. The result is a limb whose behavior can be set mechanically without changing the basic design.
The study places several oscillators on a lightweight icosahedral frame to create hopping robots. One design uses five units arranged around the base and powered through a single line at 610 kPa. At first the limbs move out of sync. After a short period, they settle into a repeating sequence and the robot hops. When suspended without ground contact, this pattern does not form. The interaction between the body and the surface creates the coupling that aligns the limbs.
To understand this coordination, the authors apply a Kuramoto type model. A phase in this context is a simple marker of position within a cycle, like the hand of a clock moving around a dial. A natural frequency is how quickly an oscillator would complete a cycle on its own. Kuramoto models describe how oscillators adjust their phases when linked. Here the coupling strength between two limbs depends on the angle between their orientations.
The model uses the cosine of that angle to express how strongly their cycles influence each other. With this rule, simulations match the measured phase differences for several layouts.
A second hopper uses four limbs. Two back legs sit at the base, and two front legs sit 10 mm closer to the center. That offset changes how the frame presses on the ground. Under constant pressure, the back legs move in phase, the front legs move in phase, and the two pairs move almost opposite each other. This produces forward hopping at about 0.3 BL s−1, where BL is body length. The same modeling approach captures this behavior and offers guidance for future layouts.
A final demonstration shows that the same unit supports sensing and safety logic. A crawler uses four units wired differently. One acts as a sensor pressed against the surface so that the robot’s weight closes its flow path. One acts as a safety valve. Two act as limbs. While on the surface, the sensor stays closed, the safety valve stays open, and the limbs oscillate. When the crawler reaches an edge, the sensor loses contact and opens. That pressure change closes the safety valve. Air stops flowing to the limbs, and the robot halts.
The paper concludes that multifunctional fluidic units reduce the number of components needed for rich behavior in soft robots. Instead of separate devices for sensing, actuation, switching, and oscillation, one building block can handle all roles depending on how it is attached. This allows behaviors to emerge from the robot’s structure and interactions rather than from external controllers.
The study also notes limits. The present unit requires several parts and careful assembly. More integrated or modular versions could simplify construction. Scaling down raises fluid flow challenges because smaller systems operate at lower Reynolds numbers, where viscous effects dominate. Past work on microvalves suggests that control is still possible in that range, though achieving strong coupling among many units may require new designs. The work points toward soft robots in which the body carries more of the sensing and control load, reducing reliance on external electronics.
For authors and communications departmentsclick to open
Lay summary
Prefilled posts
Nanowerk Newsletter
Get our Nanotechnology Spotlight updates to your inbox!
Thank you!
You have successfully joined our subscriber list.
Become a Spotlight guest author! Join our large and growing group of guest contributors. Have you just published a scientific paper or have other exciting developments to share with the nanotechnology community? Here is how to publish on nanowerk.com.
