Session F13: Biological Fluid Dynamics: Locomotion III

Surface swimming dynamics of centipedes

Terrestrial multi-legged animals such as centipedes navigate diverse environments by generating limb-stepping waves and traveling body waves. However, less is known about how these animals navigate fluid environments. Here, we challenged Lithobius foficatus (N=8, 14 leg pairs) to locomote on water. In addition, we used a Schlieren method to reconstruct and observe waves generated by the centipedes. When kept afloat by surface tension, the animals swam successfully; any instances of submersion resulted in the animal sinking without generating motion. Surprisingly, unlike on solid terrain, centipedes propelled via waves of horizontal plane body curvature propagating in the direction of motion. Using body undulation, the centipedes achieved speeds of 0.22±0.03 body lengths per gait cycle using body waves with an amplitude of 3.9±1.5 cm^-1 and 1.3±0.23 waves along their bodies. When the centipedes moved their limbs but did not use body undulation, they achieved negligible displacement. This suggests that surface swimming in this species is facilitated by the propagation of body waves, not limb flexion. We posit these direct body waves enable the animal to swim by varying the animal’s drag anisotropy (ratio of local perpendicular to parallel forces).   

Wake Dynamics and Thrust Generation of a Foil Flapping Near the Sedimentary Bed

The active flapping of airfoil generates significant wake vortex shedding and thrust force, which provides key insights for understanding the locomotion of marine animals. In this work, experimental efforts with particle image velocimetry and force sensor were conducted to characterize the wake dynamics and thrust generation of a foil flapping near the sedimentary bed, which mimics the swimming of fishes close to the seabed. Overall, the thrust generation is dominated by the Strouhal number defined by the flapping frequency, amplitudes and incoming flow velocity. Results show that with the growing flapping frequency and amplitudes, the difference of thrust generation between solid and sedimentary bed increases. The stronger flapping intensity results in more distinctive sediment suspension, where the distribution of suspended sediments is highly correlated to the wake vortex structures. The characterization of wake statistics shows that at the same flapping frequency and amplitude, cases with solid bed produces stronger ‘jet-style’ wake velocity distributions compared to the sedimentary bed counterparts.   

Not Participating

Flexible structures are ubiquitous in natural propulsors. However, the significance of fluid-structure interaction for bio-propulsion is not well understood. The relative performance of rigid and flexible fins has been investigated with an experimental trajectory optimization scheme. In an oil tank facility, a robotic fin controlled via a spherical parallel manipulator (SPM) optimizes kinematics with an automated covariance matrix adaptation evolutionary strategy (CMA-ES). This robot enables exploration of complex 3D trajectories with larger rotations than are available to natural propulsors. Previous work has shown that uniformly flexible fins converge to the same trajectories as stiff fins (with lower efficiencies) when optimized for side-force. The present study aims to compare the dynamics of rigid and flexible propulsors with thrust-optimized trajectories. Subsequent flow field measurements will be obtained using digital particle image velocimetry (DPIV) to directly compare the resulting vortex dynamics.   

Helical tail robot to study interactions at low Reynolds number

Some bacteria use helical flagella to self-propel. Their swimming dynamics are relatively well understood, but several aspects remain to be explored.  For instance, it is not clear how swimmers interact with each other especially while swimming in complex media. Instead of using actual living organisms, we have developed  an autonomous swimming robot with a helical tail that operates in the Stokes regime. The robot uses a battery-based power system, with a miniature integrated circuit design which remotely controls the rotational speed of the helical tail.  This torque-free neutrally-buoyant design mimics the swimming strategy of bacteria more closely than other previously used designs. The robot is several centimeters long; hence, we use highly viscous fluids to match the Reynolds number  to be Re ≈ 0.1.  Measurements are conducted  for a range of helical wavelengths, λ, radii, R, lengths, L, and rotation rates, ω. We provide comparisons of the experimental measurements with the predictions of the resistive force theory. Preliminary results of the interaction of the swimmer with walls, free surfaces and other swimmers are shown and discussed.   

Not Participating

Beating of a long, thin cylindrical tail is a common means of locomotion at small scales. Here, the implications of plasticity in the ambient fluid are considered. Building on some previous work in which classical slender-body theory was generalised to describe motion in a viscoplastic fluid, solutions are presented to show how a yield stress affects the motion of a cylindical swimmer. Numerical solutions span the range of the two key parameters of the problem: the wave amplitude (relative to wavelenth) and a dimensionless measure of the yield stress; these are suplemented by discussion of some limits of the problem where analytical progress is possible. When the yield stress is large, 'burrowing' is possible at sufficiently high wave amplitudes, in which the swimmer moves along its centreline at the wave speed. Conversely, at low amplitudes the swimmer can be retarded by the ambient yield stress and, for some waveforms, even move backwards.     

Rotation and polarisation in the locomotion of confined zebrafish

We investigate the individual and collective locomotion of zebrafish in a confined circular shallow environment. Experimentally, we observe that as they grow from larvae to adults, individual zebrafish transition from exploratory swimming to a regime where they tend to follow the wall. When several adults are present, the collective rotation displays significant amount of polarisation: a majority of fish rotate in the same direction. Inspired by the specific intermittent gait adopted by the fish, so-called burst-and-coast dynamics, we model the assembly as purely repulsive, persistent random walkers with realistic body and gait characteristics. Our model captures the individual confinement transition observed during ontogeny. More surprisingly, and despite the absence of “à la Vicsek” alignment terms, the model also captures the collective spontaneous polarisation observed in the experiment.   

Order in fish collective motion is modified by environnement illuminance

The level of order in animal groups on the move can display a wide range of variations, from fully disorganized aggregates to regular networks. This collective motion could not be achieved if individuals were assessing their environment independently; instead, it emerges due to interactions within the group. Ability to organize thus strongly depends on disturbance of the environment, which can alter how group members perceive their neighbors. For fish schools, information collection relies on sensory mechanisms, namely vision and flow sensing. However, quantitative description of sensory thresholds leading to schooling state transition is still lacking. We show that the group structure of rummy-nose tetra (Hemigrammus rhodostomus) exhibits distinct dynamics depending on the illumination of their habitat. Free swimming assemblies of ca. 50 fish are recorded in a large tank, where illuminance level is adjusted from 0 to 900 lux. We quantify geometrical order with polarization and milling parameters, which capture global alignment and rotation: for low light exposure, we observe little to no order; intensity of the collective motion then increases with illumination, until a threshold. Our data suggest that vision capability plays a major role in the level of order of a fish school.