Soft bodied organisms such as annelids may exploit body compliance by using their hydrostatic skeletons and muscles to burrow in granular substrates. The prevalence and performance of soft structures in biology has inspired researchers to incorporate soft materials into new robotic systems with adaptive and robust qualities. In this work, we investigate the design of soft digging robots inspired by the bristled worm, (polychaetas).

The behavior of soft structures in granular environments is complex and still not well understood. We detail the experiments, design, and fabrication of a soft robotic system capable of maneuvering in granular substrates and investigate actuation strategies for drag reduction inspired by the bristled worm's biomechanical behaviors. The soft robotic system is composed of three main actuator segments, with the leading segment being the focus of interest for analysis of this complex locomotion. We implemented and studied two methods of actuation in our soft-robot: peristaltic expansion and bi-directional bending.

We compared the drag force experienced by the leading segments that reproduce these active strategies to the force experience by rigid, and unactuated, soft versions. We find that biomechanical behaviors can have a significant impact on locomotion strategies in granular substrates. Based on these results, we demonstrate a tethered, three-segment soft robot capable of digging through granular media. In summary, we find that over a range of movement speeds, soft-robots performing peristaltic expansion at their tip experience the least drag force. Soft-robots with unactuated tips experienced the largest drag resistance emphasizing the importance of controlling the tip stiffness to enable effective subsurface movement.