ScaFi: length-scalable, compliant, parametric robotic fish design for operation in multiple environmental niches

Abstract Aquatic creatures provide inspiration for robotic design, their morphologies spanning a vast range of scales. These scaled forms are suited to the performance and environmental demands of their respective habitats, influencing characteristics like speed and tail beat frequency. However, in artificial systems, the design of a single robot that can operate effectively across a wide range of physical sizes remains a significant challenge, with current tethered or untethered designs lacking generalizability. To address this, we propose a design of a compliant robot fish, ScaFi, which requires only a single motor and can scale from several tens of centimeters to a few meters in body length. The tail design, inspired by subcarangiform swimmers, consists of a rigid front-end and a compliant tail with fiberglass rods and a 'cross-over' tendon routing. By applying a scaling law derived from tail deflection, we can ensure a consistent kinematic behavior across scales, allowing the robot to generate a bio-inspired wake regardless of its size. The primary goal of this work is to develop a parametrically length-scalable design for a compliant robotic fish that can be deployed in diverse environmental niches. We demonstrate the scalability of the design in three physical prototypes that span lengths from 50 cm to over 2.5 meters. Particle image velocimetry tests confirm that the motion generates bio-inspired vortex wakes, while swimming experiments and field deployments demonstrate the practical utility of a multi-scale design and its ability to adapt to diverse aquatic conditions.