Vertical slot fishways are a crucial measure to mitigate the blockage of fish migration caused by hydraulic engineering infrastructures, but their passage efficiency is often hindered by the complex interactions between fish behavior and hydrodynamic conditions. This study combines computational fluid dynamics (CFD) simulations with behavioral laboratory experiments to identify the hydrodynamic characteristics and swimming strategies of three types of fishways—Central Orifice Vertical Slot (COVS), Standard Vertical Slot (SVS), and L-shaped Vertical Slot (LVS)—using the endangered species Schizothorax prenanti from the upper Yangtze River as the study subject. The results revealed that (1) a symmetric and stable flow field was formed in the COVS structure, yet the passage ratio was the lowest (50%); in the SVS structure, high turbulent kinetic energy (peak of 0.03 m²/s²) was generated, leading to a significant increase in the fish’s tail-beat angle and amplitude (p < 0.01), with the passage time extending to 10.2 s. (2) The LVS structure induced a controlled vortex formation and created a reflux zone with low turbulent kinetic energy, facilitating a “wait-and-surge” strategy, which resulted in the highest passage ratio (70%) and the shortest passage time (6.1 s). (3) Correlation analysis revealed that flow velocity was significantly positively correlated with absolute swimming speed (r = 0.80), turbulent kinetic energy, and tail-beat parameters (r > 0.68). The LVS structure achieved the highest passage ratio and shortest transit time for Schizothorax prenanti, demonstrating its superior functionality for upstream migration. This design balances hydrodynamic complexity with low-turbulence refuge zones, providing a practical solution for eco-friendly fishways.

The efficient propulsion of flexible swimmers in unsteady flow is a fundamental problem in fluid–structure interaction (FSI), governed by the physical trade-off between maximizing propulsive efficiency and maintaining hydrodynamic stability. This study integrates high-fidelity computational fluid dynamics simulations with high-resolution kinematic tracking to quantify the response mechanisms of four biological prototypes with distinct geometric characteristics in a controlled Kármán vortex street. Our results reveal a morphology-driven spectrum of interaction modes comprising three distinct physical regimes: Vortex synchronization (efficiency-optimization), low-vorticity navigation (stability-prioritization), and hydrodynamic instability (physical failure). In the synchronization regime, we quantify a characteristic non-linear saturation in tail-beat frequency. This kinematic signature is consistent with a flow-induced resonance state, where phase-locking maximizes momentum extraction from the environment. In contrast, laterally compressed bluff-body prototypes are highly susceptible to destabilizing yawing moments, necessitating a spatial avoidance strategy to mitigate the risk of biomechanical failure. This work quantifies how geometric constraints dictate the physical boundary conditions for the FSI mechanism, providing a critical mechanical framework for the multi-mode adaptive control of bio-inspired vehicles in complex turbulent environments.

The interaction between fish and complex flow environments has drawn attention; however, the underlying mechanisms remain poorly understood. This study investigated the swimming behavior of Aristichthys nobilis (A.N.) in the wake regions formed behind rigid vertical bluff bodies with four different cross-sectional shapes, under flow velocities ranging from 0.1 to 0.7 m/s. Key fish swimming kinematic parameters, including swimming speed, tail-beat frequency, amplitude, and angle, were quantified. The results showed that A.N. preferentially performed Kármán gaits in the wakes of bluff bodies with circular and rhombic cross sections. In these cases, the occurrence probabilities of the Kármán gait ranged from 72% to 55%, significantly higher than those observed for square or rectangular cross sections. Moreover, at lower flow velocities (0.1–0.3 m/s), the predominant swimming mode was gliding, characterized by lower swimming speed and tail-beat frequency, while at higher velocities, A.N. tended to switch to the Kármán gait, likely to enhance swimming efficiency.

Complex unsteady flow environments present substantial challenges to the stability and energy efficiency of autonomous underwater vehicles (AUVs). While fish can utilize mechanisms like the Kármán gait (KG) to conserve energy, a quantitative understanding of the underlying wave mechanics and morphological limitations remains lacking. This study conducts a comparative kinematic analysis of two morphologically distinct species—the streamlined Leuciscus baicalensis and the deep-bodied Carassius auratus gibelio—in the wake of a D-shaped cylinder. Using the two-dimensional Hilbert transform for wave decomposition and time-series analysis, we quantify body wave evolution and intermittent swimming across uniform flow (UF), entrainment (EN), and KG regimes. Results reveal distinct physical modes: UF relies on active propulsion, EN exhibits active damping with recoil noise, whereas KG operates as a passive resonance mode where hydrodynamic loading forms standing waves. Temporal analysis identifies a gait-dependent control shift from stability-driven intermittent bursting in shear layers—where intermittency correlates positively with velocity—to hydrodynamic lock-in with a 100% duty cycle in vortex streets. Critically, morphological compliance dictates the stability boundary: the streamlined species sustains resonance at 0.8 m/s, whereas the deep-bodied species experiences control divergence due to excessive transverse added mass. These findings provide key criteria for a dual-mode AUV design.

How flexible swimmers exploit unsteady flows is a longstanding question in fluid–structure interaction. Kármán gait allows fish to hold station in vortex wakes at reduced energetic cost, yet how the wake organizes and stabilizes this gait remains unclear. Here, we combine comparative experiments on live and dead fish with passive bionic models, hydrodynamic reconstruction, and reduced-order modelling. Across 20 species, Kármán gait exhibited a common phase-locked organization in which wake–body interactions shaped much of the body-center and tail motion, while active regulation was concentrated in the anterior body. A forced damped-oscillator model captured the observed phase lag and near-resonant response. Bionic-model experiments identified sufficient compliance and wake–body frequency matching as mechanical constraints, whereas live-fish observations showed that stable entrainment was concentrated at an intermediate body–wake scale. These results identify Kármán gait as a physically gated, wake-forced state, with implications for fish ecology and bioinspired robotics.