Computationally assisted design and selection of maneuverable biological walking machines

ABSTRACT The intriguing opportunities enabled by the use of living components in biological machines have spurred the development of a variety of muscle-powered bio-hybrid robots in recent years. Among them, several generations of bio-hybrid walkers have been established as reliable platforms to study untethered locomotion. However, despite these advances, such technology is not mature yet, and major challenges remain. This study takes steps to address two of them: the lack of systematic design approaches, common to bio-hybrid robotics in general, and in the case of bio-hybrid walkers specifically, the lack of maneuverability. We then present here a dual-ring biobot, computationally designed and selected to exhibit robust forward motion and rotational steering. This dual-ring biobot consists of two independent muscle actuators and a 4-legged scaffold asymmetric in the fore/aft direction. The integration of multiple muscles within its body architecture, combined with differential electrical stimulation, allows the robot to maneuver. The dual-ring robot design is then fabricated and experimentally tested, confirming computational predictions and turning abilities. Overall, our design approach based on modeling, simulation, and fabrication exemplified in this robot represents a route to efficiently engineer biological machines with adaptive functionalities.