Ten-cent soft robots: Oxford engineers build flexible actuators from vacuum bags and a laser cutter

A soft robotic gripper that lifts 25 times its own weight. Crawling robots. Swimming robots. Inflatable turtles and cranes. All made from materials costing less than a dime.

Engineers at the University of Oxford have published a fabrication method in Advanced Science that strips soft robotics down to three components: commercial thermoplastic vacuum pouches, a standard vacuum sealing machine, and a laser cutter. The total material cost per actuator is under $0.10. Fabrication time: less than 10 minutes.

Cut, seal, inflate

Soft robotic actuators are the flexible components that power movement in soft robots. They bend, twist, or extend when pressurized with air. Conventional methods for making them typically involve silicone molding, specialized 3D printers, or complex textile lamination -- processes that are slow, expensive, and require equipment most labs do not have.

The Oxford approach sidesteps all of that. The team, led by Professor Antonio Forte and postdoctoral researcher Ashkan Rezanejad, starts with commercially available vacuum-sealable plastic pouches. By removing air between layers using a standard vacuum sealer before laser processing, they simultaneously seal and shape inflatable structures in a single step. The laser cutter defines the geometry; the vacuum seal creates the airtight chamber.

The result is a programmable bending actuator. When air is pumped in, it bends predictably based on the cut geometry. By adjusting the geometric parameters of the laser cut, engineers can control the direction, magnitude, and shape of the bend -- producing spirals, curves, or letter-shaped structures.

Performance that holds up

Cheap and fast fabrication means nothing if the products fail under use. The Oxford team ran systematic durability tests, subjecting their thermoplastic actuators to 100,000 inflation-deflation cycles. The structures survived.

The actuators also generated strong output forces at relatively low air pressures, which matters for practical applications where high-pressure pneumatic systems may not be available or desirable. The gripper the team built could lift 25 times its own weight -- a performance metric that compares well with actuators made using far more expensive methods.

The team also developed a computational design framework that lets engineers simulate how different cut geometries will behave before fabrication. This turns the design process from trial-and-error into something predictable, which is important for scaling the approach beyond one-off prototypes.

Who this is for

The immediate beneficiaries are researchers and educators who want to experiment with soft robotics without investing in specialized fabrication equipment. University labs, start-ups, and classroom settings can all access vacuum sealers and laser cutters.

"By lowering the financial and technical barriers to fabrication, this advance could significantly democratise and accelerate soft robotics research and prototyping," Forte said.

Rezanejad pointed to the educational potential: "We even produced inflatable animal structures, including turtles and cranes. By enabling creative and artistic projects, our method could be particularly valuable for education and attracting students to soft robotics."

What it cannot do -- yet

The method produces bending actuators. More complex motions -- twisting, multi-directional movement, or the kind of fine manipulation required for surgical applications -- are not yet achievable with this approach. The team has identified these as goals for future work, along with testing other compatible thermoplastic materials that might offer different mechanical properties.

The actuators are made from plastic pouches designed for food storage, which raises questions about long-term material degradation, especially in demanding environments like underwater operation or continuous industrial use. The 100,000-cycle durability test is encouraging, but it was conducted under controlled laboratory conditions.

Soft robotics as a field is exploring applications in minimally invasive medical devices, wearable technologies, and hazardous-environment exploration. Whether vacuum-bag actuators can compete with more sophisticated (and expensive) fabrication methods in these demanding applications remains to be demonstrated. The value of the Oxford method may lie less in replacing existing approaches and more in enabling rapid prototyping and iteration before committing to costlier production methods.