Soft robots, constructed from pliable materials, are particularly suited for activities such as traversing irregular surfaces or manipulating fragile items.
A primary objective in the field of soft robotics is to integrate behavior and decision-making capabilities directly into the robot's physical design, allowing for machines that are more adaptable and responsive.
This type of autonomous behavior, which arises from interactions between the body and its environment, is often challenging to replicate using conventional electronic circuits that require intricate sensing, programming, and control mechanisms.
Just as fireflies can begin flashing in unison after watching one another, the robot’s air-powered limbs also fall into rhythm, but in this case through physical contact with the ground rather than visual cues.
This emergent behavior has previously been observed in nature, and this new study represents a major step forward towards programmable, self-intelligent robots.
Dr. Mostafa Mousa, Study Lead Author, Department of Engineering Science, University of Oxford
Their primary innovation was the creation of a compact, modular component that uses air pressure to execute mechanical tasks, much like an electronic circuit operates using electrical current.
Depending on its configuration, this singular block can either: Activate (move or change shape) in reaction to variations in air pressure, operating similarly to a muscle; detect pressure variations or contact, akin to a touch sensor; or toggle air flow between ON and OFF states, comparable to a valve or a logic gate.
Much like LEGO components, several identical units (each measuring a few centimeters) can be assembled to create various robots while maintaining the fundamental hardware design.
In the research, the scientists built tabletop robots (approximately the dimensions of a shoebox) that were capable of hopping, shaking, or crawling.
In a specific configuration, the researchers discovered that each individual unit is capable of automatically integrating all three roles simultaneously, allowing it to produce rhythmic movement independently as soon as constant pressure is exerted.
When multiple responsive units are interconnected, the movements start to synchronize naturally, without the need for any computer control or programming.
These behaviors were employed to construct a shaker robot (capable of sorting beads into various containers by tilting a rotating platform) and a crawler robot (which could sense the edge of a table and autonomously halt, thereby preventing a fall).
In both instances, the synchronized movements were accomplished solely through mechanical means, without any external electronic control.
The synchronized behavior is observed solely when the robots are interconnected and in contact with the ground.
The researchers used a mathematical framework known as the Kuramoto model, which illustrates how networks of oscillators can achieve synchronization, to elucidate this behavior.
Encoding decision-making and behavior directly into the robot’s physical structure could lead to adaptive, responsive machines that don’t need software to ‘think.’ It is a shift from ‘robots with brains’ to ‘robots that are their own brains.’
That makes them faster, more efficient, and potentially better at interacting with unpredictable environments.
Antonio Forte, Study Co-author and Professor, Department of Engineering Science, University of Oxford
This demonstrated that intricate, synchronized movement can arise in the robots solely due to their physical configuration when they are mechanically interconnected through the environment.
In this scenario, the movement of each robotic leg subtly influences the others through the common body and ground reaction forces.
This establishes a feedback loop in which the forces conveyed through friction, compression, and rebound connect the movements of the limbs, resulting in unprompted coordination.
While the soft robots that have been developed are presently at a tabletop scale, the researchers assert that the design principles are not dependent on scale.
In the near future, the researchers plan to explore these dynamic systems to create energy-efficient untethered locomotors.
This would represent a significant advancement towards the large-scale implementation of these robots in extreme environments where energy is limited and adaptability is essential.
Multifunctional Fluidic Units for Emergent, Responsive Robotic Behaviors
Watch the RADLab's video summary of the study. Video Credit: University of Oxford
Journal Reference:
Mousa, M., et al. (2025). Multifunctional Fluidic Units for Emergent, Responsive Robotic Behaviors, Advanced Materials. DOI:10.1002/adma.202510298.