Princeton prints motor-free origami robot hinges with closed-loop thermal control
Princeton engineers reported on March 20 in Advanced Functional Materials that they built a soft-rigid hybrid robot architecture using 3D-printed liquid crystal elastomer hinges, embedded flexible printed circuit boards, and temperature-feedback control, allowing electrically driven folding motion without motors or external pneumatic hardware. The central engineering result is not just a wing-flapping crane demonstration; it is a manufacturing method that embeds sensing, heating, and fold definition directly into the structure so the robot can execute programmable motion while repeatedly returning to shape with minimal visible degradation.
The Princeton team, led by Emily Davidson and Glaucio Paulino, used a layer-by-layer direct ink writing process to print molten liquid crystal elastomer into patterned regions with controlled molecular orientation. That orientation matters because the material contracts along programmed directions when heated through its phase transition. By stacking and joining zones with different alignments, the researchers created hinges that bend in predetermined ways rather than relying on separate joints, gears, or external actuators.
That is the key systems advance: actuation is no longer bolted onto an origami robot after fabrication. It is designed into the material architecture itself. Flexible printed circuit boards were embedded during printing instead of being attached in a later assembly step. In practical mechanical-design terms, that reduces process handoffs, improves alignment between heater traces and active hinges, and makes the final structure closer to a repeatable manufacturing workflow than many lab-scale soft robots assembled by casting, gluing, or over-molding.
The embedded circuits do two jobs. First, they provide localized Joule heating at selected hinges, giving the robot addressable fold control. Second, they carry temperature sensors used for closed-loop control, allowing the system to correct for the small thermal and motion errors that build up over repeated shape changes. That feedback layer is important because soft structures are harder to command precisely than rigid linkages; without sensing, repeatability usually suffers as heat spreads unevenly and materials drift.
The paper’s best-known demonstration is an origami crane whose body and wings can be actuated in sequence and whose wings flap when power is applied. The more important technical point is how that motion is achieved. Lightweight fiberglass stiffener panels placed between hinges keep most of the structure comparatively rigid, so folding is confined to the intended crease regions. In effect, the design separates compliant zones from load-bearing panels, which is a familiar requirement in deployable mechanisms but here is handled inside a printed soft-rigid composite.
The researchers also reported durability results that strengthen the manufacturability claim. Their hinge system demonstrated more than 1,500 actuation cycles with minimal performance degradation after an initial break-in period. The test was stopped because of time limits rather than confirmed failure, so the full life limit remains unknown, but 1,500-plus cycles is enough to show that the architecture is moving beyond one-off shape morphing demos and toward repeatable operation.
Speed is another useful data point. The team reported full cycle times under 100 seconds and described that as more than two times faster than comparable electrically driven liquid crystal elastomer actuators, while also being substantially faster than environmental heating methods that warm the whole structure. That improvement comes from localized heating and thermal isolation: the flexible circuit design includes large copper dead zones around folds to dissipate heat and reduce unwanted thermal coupling between neighboring hinges.
There are still clear operating boundaries. This is not a high-bandwidth replacement for motors in conventional robots, and the paper does not claim that. Thermal actuation trades response speed for structural simplicity and integration. The control system also had to limit pulse-width modulation output in some tests so hinge motion would better track temperature, which is a reminder that precise thermal actuation depends on carefully managing lag between heating, material response, and cooling.
Another meaningful addition is the software workflow. Because origami robots can become design-intensive quickly, the team developed a tool called OriCadLCE to generate fold patterns, print paths, and fabrication files for the embedded circuit layers. That matters because integrated robotics platforms often fail not on a single material or actuator, but at the interface between geometry, electronics, and manufacturing. A design pipeline that resolves hinge overlaps, embedded-sheet placement, and fabrication outputs is a practical contribution, not just supporting software.
The work began as an undergraduate thesis by David Bershadsky, who graduated from Princeton in electrical engineering in 2024 and is now a graduate student at the University of Texas at Austin. He also released the design tool with the project dataset, which should make the result easier to reproduce and extend than many soft-robot papers that stop at a prototype description.
What Princeton has shown here is a credible path toward more integrated soft robotic mechanisms: printed smart-material hinges for motion, embedded flex circuits for actuation and sensing, and origami geometry for deployable structure. The crane is the visual proof, but the more durable contribution is the fabrication stack. In soft robotics, that kind of integration usually matters more than a single demo shape because it determines whether a concept can be built consistently, controlled repeatably, and redesigned without starting from scratch each time.
By Jonathan Barrett — Editor for AMI’s future mobility and autonomous systems section, with two decades covering robotics, e-mobility, drone-vehicle convergence, and transport mechanical systems.
