SLIDER robot achieves 21.7 cm/s swimming via model-based gait optimization

A team of researchers from Florida State University, Vanderbilt University, and the University of South Florida introduced SLIDER (Soft Lamprey-Inspired Dual Environment Robot), a soft robotic system designed for both swimming and climbing. Their paper, published on arXiv, presents a comprehensive model-based optimization framework that combines Lighthill's large-amplitude elongated body theory for fluid dynamics with a geometrically and materially nonlinear structural model. The fluid-structure interaction equations are solved implicitly using an efficient second-order box method, enabling fast computation. A pneumatic manifold system actuates SLIDER in a quiescent water tank, allowing validation against experiments. Using a genetic algorithm, the team co-optimized the swimming gait pattern and caudal fin design, taking into account the robot's climbing morphology. They achieved a tethered swimming speed of 21.7 ± 0.4 cm/s, equivalent to 0.59 body lengths per second. The study also reveals that low-frequency swimming is dominated by resistive environmental forces, while high-frequency swimming is primarily affected by inertial fluid forces. This work not only advances the design of soft robotic swimmers but also demonstrates a transferable optimization pipeline for multimodal robots that need to operate in both aquatic and terrestrial environments, opening new possibilities for exploration, inspection, and rescue missions.