Scientists from the University of California, Santa Barbara, and the Dresden University of Technology have developed a robotic material that can shapeshift into almost any form and carry many times its weight.
The UC Santa Barbara and TU Dresden team studied living embryonic tissues‘ ability to alter mechanical properties through coordinated cellular behavior, translating that capability into gears, photoreceptors, and rolling magnets.
The advancement consists of a cohesive network of individual robots working as a single adaptive structure by precisely controlling force fluctuations and polarity to alter the rigidity and fluidity of the robotic cells.
Robotic Materials After Living Cells
Recent developments involving shapeshifting robot technologies have been referred to by names like “claytronics” and “programmable matter.” However, robotic systems that can change shape on the fly, allowing them to adapt to needs that aren’t envisioned at the time of production while still maintaining cohesion, have remained elusive. Previous designs have suffered limitations, although some instances have been capable of reconfiguring at will despite remaining incapable of bearing loads. Others, by contrast, can bear loads but must break cohesion to reconfigure.
The international team behind the recent achievement settled on embryonic cells when seeking natural models for the behavior they sought to develop in their robotic material. At the first stages of life, cells can reshape, rearrange, repair, and regenerate as organisms develop.
Two cellular processes, in particular, caught the team’s eye: tissue fluidization, in which tissue switches into a fluid-like state, and convergent elongations, where polarized cells stretch out. Mimicking those cellular processes, the team designed a robotic collective capable of controlling local object rearrangements across the entire system.
A Mass Of Cells as Robotic Material
The robotic material copies embryonic cells‘ ability to stick together, apply pressure to one another to move cells within the mass, and, crucially, coordinate these activities within the collective. Each individual unit has controls for polarity, force, and adhesion to allow it to behave in the desired cell-like manner.
Eight partially exposed gears allow the robotic material cells to push against each other to reconfigure more quickly. Beyond that, photoreceptors enable the cells to receive light-based signals, and magnets around their perimeter create the necessary adhesion.
With the system designed, the team developed a mathematical model for studying the emerging behaviors in robotic material collective. Between simulations and a small four-unit test model, the team identified the necessary forces and polarities to effectively control the cluster as it transformed from fluid to solid states and from one structure into another.
The researchers then used a small, three-block unit to test how much pressure the system could support from an exterior load. An interesting discovery in these real-world tests was that power fluctuations enabled a power-limited system to perform tasks outside its normal ability, increasing the collective’s rearrangement and load-bearing capabilities.
Demonstrating From Fluid to Solid
The team subjected the robots to over 200 hours of testing and characterization, finding the robotic material so robust that it could perform as normal even with up to 28% of the collective’s motors broken.
The robot units morphed from separate pillars into a stable load-bearing arch in a demonstration. Beyond its initial formation, the collective could heal structural defects and apply force to move items. In another example of manipulating objects, the system could flow around an object and rigidify into a wrench shape, allowing it to exert torque.
The tests also revealed usage at the practical scale, where the robot units successfully attained a configuration comparable to the size of a human before transitioning to a fluid state.
The team’s findings are detailed in a paper, “Material-like Robotic Collectives with Spatiotemporal Control of Strength and Shape,” which appeared on February 21, 2025, in Science.
Ryan Whalen covers science and technology for The Debrief. He holds an MA in History and a Master of Library and Information Science with a certificate in Data Science. He can be contacted at ryan@thedebrief.org, and follow him on Twitter @mdntwvlf.