Biologists and engineers at the University of California, Berkeley have developed a robot named Salto, which can mimic a squirrel's ability to land on narrow branches. This innovation is based on research into the biomechanics of squirrels, focusing on their leaps and landings. The study was published in the journal Science Robotics.

Robert Full, one of the senior authors of the paper and a professor at UC Berkeley, highlighted the significance of this advancement. He stated that while current robots are functional, they lack the agility to navigate challenging environments like squirrels do. "The robots we have now are OK, but how do you take it to the next level?" he asked.

Justin Yim, a former graduate student at UC Berkeley and co-first author of the paper, played a key role in adapting what was learned from squirrels to Salto. The robot was originally developed in 2016 by Ronald Fearing's lab at UC Berkeley's Department of Electrical Engineering and Computer Sciences (EECS). Yim explained how he programmed Salto to mimic human strategies for maintaining balance during jumps.

Using these insights, Yim plans to work on a NASA-funded project involving a small robot capable of exploring Enceladus, one of Saturn’s moons. This new design is informed by biomechanical analysis detailed in another paper accepted for publication in the Journal of Experimental Biology.

The development involved collaboration between biology students from Full’s Polypedal Lab and engineering students from Fearing’s Biomimetic Millisystems Lab. The team discovered that squirrels absorb most landing impact with their front legs and adjust their braking force for balance—a technique they incorporated into Salto.

Yim also redesigned Salto with adjustable leg forces alongside its existing reaction wheel torque system. This allowed Salto to achieve balanced landings without gripping capabilities.

Future work could explore more advanced grippers for expanded landing abilities. Full is investigating further into how squirrels manage torque without thumbs for prehensile grasping—a potential advantage when evading predators quickly.

Salto’s development demonstrates significant potential for single-legged robots due to concentrated power efficiency during high jumps. According to Yim: "One leg is the best number for jumping."

Co-authors include Ronald Fearing; Eric Wang from MIT; Nathaniel Hunt from University Nebraska Omaha; Hannah Stuart from UC Berkeley Mechanical Engineering; Stanley Wang; Duyi Kuang—all contributing valuable insights towards this innovative research funded by U.S Army Research Office & National Institutes Health grants.