Photo credit: www.sciencedaily.com

Revolutionizing Robotics: A New Age of Smart Materials

A collaboration between researchers at UC Santa Barbara and TU Dresden is redefining the intersection of robotics and material sciences, crafting a novel collective of robots that exhibits behaviors inspired by biological systems.

“We’ve developed a method to enable robots to behave more like a cohesive material,” stated Matthew Devlin, a former doctoral researcher under mechanical engineering professor Elliot Hawkes at UCSB, and the principal author of a study published in the journal Science. This innovative approach comprises disk-shaped autonomous robots resembling small hockey pucks, which can autonomously assemble into varying structures, each with distinct material properties.

One of the team’s primary objectives was to engineer a robotic material that possesses both rigidity and the ability to flow when shape alterations are required. As Hawkes emphasized, the ideal scenario involves robots that don’t merely react to external forces but rather respond to internal cues, allowing them to adopt and maintain various shapes while also transitioning into new ones when needed.

For design inspiration, the researchers leaned on the work of Otger Campàs, a former professor at UCSB and now the director of the Physics of Life Excellence Cluster at TU Dresden, who studies the physical shaping mechanisms of embryos. “Living embryonic tissues exemplify the ultimate smart materials,” he noted, highlighting their innate abilities to self-shape, self-repair, and modulate their material strength dynamically. During his time at UCSB, Campàs discovered that embryonic cells could fluidize their structures akin to glass. “To modify their shape, these cells facilitate transitions between solid and fluid phases—a phenomenon termed rigidity transitions in physics,” he added.

During embryogenesis, cells exhibit an extraordinary capacity to organize themselves, transforming from an undifferentiated mass to a complex organism characterized by distinct shapes and various material properties such as bones and nervous tissue. The researchers focused on three critical biological processes that enable these rigidity transitions: the active forces that cells exert on one another, the biochemical signals coordinating their movements, and their adhesive properties which confer the final structure’s stiffness.

In the context of robotics, the forces analogous to intracellular movements manifest as inter-unit tangential forces. This interaction is facilitated by eight motorized gears positioned around each robot’s edge, enabling them to interact and maneuver even in dense formations.

Biochemical signaling is simulated in the robots through a global orienting system. As Hawkes explained, “Each robot understands its orientation, allowing it to apply the correct forces.” This coordination enables the collective of robots to reconfigure, much like cells elongating to shape an organism.

The robots achieve this through light sensors equipped with polarized filters. When illuminated, the polarizing light informs each robot on how to adjust its gears, enabling collective movements. “By exposing them to a consistent light field, we can instruct them all at once, allowing them to align and execute the desired arrangement,” Devlin elaborated.

To simulate cell adhesion, the units utilize magnets that can be switched on to attract neighboring robots. During tests, researchers discovered that varying signal patterns significantly influenced the robots’ ability to form necessary configurations. “We’ve previously established that fluctuations in forces among cells are crucial for the transition from solid-like to fluid-like states in living tissue. Thus, we incorporated these force fluctuations into our robots,” stated Campàs.

The interplay between signal fluctuations and inter-unit forces dictates whether the robotic collective remains rigid or operates more fluidly. “Essentially, enhancing both fluctuations and forces yields a material that flows more readily,” Devlin noted. This dynamic property allows the robot assembly to transition between shapes, with the ability to regain rigidity once the force fluctuations are disabled.

Significantly, the research revealed that these signal fluctuations enable the collective to adapt its shape and strength using less energy than if these signals were perpetually active, a noteworthy finding that emerged during data collection. Hawkes highlighted the importance of this for designing energy-efficient robots.

The researchers successfully engineered the robot collective to function as a smart material, with sections dynamically activating forces to fluidize while others maintain rigidity. This modulation enabled the development of robotic materials capable of bearing substantial loads, reshaping, manipulating objects, and even exhibiting self-healing characteristics.

Currently, the experimental robot collective consists of 20 larger units, though simulations carried out by Sangwoo Kim, a former postdoctoral fellow in Campàs’s laboratory now at EPFL, suggest that this system could be scaled to accommodate larger, miniaturized units, enhancing its material-like qualities.

The implications extend beyond robotics. The researchers posit that such robotic collectives can facilitate the exploration of phase transitions in active matter, provide insights into active mechanics within particulate systems, and potentially inform hypotheses in biological research. By integrating current control methodologies and machine learning techniques, future advancements in these robotic collectives could pave the way for new, emergent capabilities in smart materials.

Source
www.sciencedaily.com