Aquatic locomotion by an elastically mounted flexible foil actuated by an oscillating force

arXiv:2606.24314 · physics.flu-dyn · Submitted 2026-06-23 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Aquatic locomotion by an elastically mounted flexible foil actuated by an oscillating force".

Mira: Here's the gist of the paper. An analytical formulation of the fluid-structure interaction of a flexible foil driven by an oscillating force actuating on its elastically mounted leading edge, so that it can heave, pitch and deform passively with the hydrodynamic forces, is used to investigate the aquatic locomotion of a body responsible for the whole drag and thrusted by the oscillating flexible foil.

Kai: Model Development The study focuses on modeling aquatic locomotion using a virtual rigid body carrying most of the mass and drag, self-propelled by a rear oscillating foil that generates thrust.

Mira: The VB-OF model is used because it allows for an analytical solution of the flow-structure interaction.

Kai: This model differs from undulatory locomotion by modeling the oscillating foil as elastically connected to the virtual body, simulating muscle flexibility.

Title and authors: Kai: So let's talk about the title itself, "Aquatic locomotion by an elastically mounted flexible foil actuated by an oscillating force," and what it tells us right off the bat.

Mira: It immediately points to a system where flexibility and oscillation work together to drive movement in water.

Lev: And from what I can gather, it's not just about one thing happening, but how the foil deforms passively alongside the body as well.

Kai: They mention that this setup lets the foil heave, pitch, and deform passively with the hydrodynamic forces acting on it.

Mira: Which means they are exploring a much broader range of motion than if you just had a rigid plate oscillating in place.

The paper's summary: Kai: The paper summarizes their model as using this virtual rigid body carrying most of the mass and drag, which is then self-propelled by a rear oscillating foil that generates the thrust needed for locomotion.

Mira: They use what they call the VB-OF model, and they emphasize that this specific setup is different from undulatory locomotion you see in many real fish.

Lev: Undulatory motion involves a traveling wave along the whole body or fins to push fluid backward, whereas this paper models a distinct mechanism.

Kai: This model allows for passive heave and pitch plus the passive flexural deformation of the foil itself, all driven by an oscillating force that mimics muscle action.

Mira: And they use a small displacement approximation to derive the equations for those forces and moments by integrating them along the chord length of that foil.

The paper's improvements: Kai: The paper suggests that including this passive heave and deformation opens up a much wider parametric range when searching for optimal self-propulsion conditions.

Mira: They found a new local optimum condition near the translational spring branch of the elastic support resonance, which happens at lower stiffness values.

Lev: That's interesting because it suggests that for certain conditions, being less rigid isn't necessarily bad for performance; it actually helps find a better spot on the performance curve.

Kai: They also found that the best propulsion performance occurs near the translational spring branch when the non-dimensional frequency and stiffness values are small.

Mira: So, they’re saying that this flexibility really enhances stride length and efficiency associated with that translational spring mode for both small frequencies and stiffnesses.

Conclusion: Kai: So wrapping up, the main point is that this analytical model helps characterize the best self-propulsion performance for any set of given parameters, provided those resulting pitch, heave, and deformation amplitudes stay small.

Mira: They really highlight how including passive heave and deformation expands the parametric space available to search analytically for better locomotion conditions.

Lev: The conclusion is that the best propulsion performance is found near the translational spring branch for small non-dimensional frequency and stiffness values.

Kai: It points towards a not too rigid foil being optimal when you're looking at small non-dimensional frequency and stiffness values located around that translational branch of the springs resonance.

Mira: They also state that this flexibility lets you get the same thrust with smaller pitch and heave amplitudes than you would for a rigid foil, even up to an amplitude of one point two eight on that foil.

Lev: From an engineering standpoint, it's valuable because it gives them a solid approximation to analyze these fish-like robots, even for those larger heave amplitudes.

Kai: So the paper with this title shows that by adding flexibility, you can explore a better region for optimal locomotion conditions around the translational spring branch of the springs resonance.

R. Fernandez-Feria

Fluid Mechanics Group, IMEC, University of Malaga

physics.flu-dyn

Submitted: 2026-06-23

Updated: 2026-06-23

Comments: 23 pages, 9 figures

Journal ref: J. Fluids Structures, 148, 104750 (2026)

DOI: 10.1016/j.jfluidstructs.2026.104750

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 83/100

The gist: An analytical formulation of the fluid-structure interaction of a flexible foil driven by an oscillating force actuating on its elastically mounted leading edge, so that it can heave…

Key concepts

VB-OF Model
This model treats the aquatic body as a virtual rigid body that is self-propelled by a rear oscillating foil. It is used because it enables an analytical solution for how the fluid forces interact with this structure, simulating muscle flexibility in the foil.
Euler-Bernoulli Beam Equation
This equation describes the bending of the flexible foil along its chord length. By solving this equation using a small displacement approximation, researchers can derive the fluid forces and moments acting on the foil based on its deformation.
Non-dimensional Parameters
These are parameters like frequency and stiffness that are scaled to make equations simpler and easier to solve. Redefining them with a 'hat' symbol simplifies the system of linear equations, allowing for an iterative process to find optimal performance values.
Translational Spring Branch
This specific branch in the spring resonance analysis is identified as a location where optimal self-propulsion performance occurs. The study found that better locomotion is achieved when operating near this branch, especially when the foil's stiffness is not too rigid.

Terminology

Summary

An analytical formulation of the fluid-structure interaction of a flexible foil driven by an oscillating force actuating on its elastically mounted leading edge, so that it can heave, pitch and deform passively with the hydrodynamic forces, is used to investigate the aquatic locomotion of a body responsible for the whole drag and thrusted by the oscillating flexible foil. p. 0

Where the paper says it

Fernandez-Feria1† 1Fluid Mechanics Group, IMEC. UMA, University of Malaga, Dr Ortiz Ramos s/n, 29071 M ´ alaga, Spain ´ (Received xx; revised xx; accepted xx) An analytical formulation of the fluid-structure interaction of a flexible foil driven by an oscillating force actuating on its elastically mounted leading edge, so that it can heave, pitch and deform passively with the hydrodynamic forces, is used to investigate the aquatic locomotion of a body, responsible for the whole drag and thrusted by the oscillating flexible foil.

Model Development

The study focuses on modeling aquatic locomotion using a virtual rigid body carrying most of the mass and drag, self-propelled by a rear oscillating foil that generates thrust. The VB-OF model is used because it allows for an analytical solution of the flow-structure interaction. p. 1

Where the paper says it

2021; Graziani et al. 2025, 2026), specially when this model, which for short is designated VB-OF model (for virtual body-oscillating foil), allows for an analytical solution of the flow-structure interaction.

This model differs from undulatory locomotion by modeling the oscillating foil as elastically connected to the virtual body, simulating muscle flexibility. p. 1
Where the paper says it

But aquatic animals whose locomotion mode approaches that of the VB-OF model, such as tunas and bonitos, are among the fastest and most efficient swimmers in sustained cruise (Webb 1984), and they are endowed by nature with a long, almost rigid-fin tail, thus approximating better the further assumption of a two-dimensional foil used in most theoretical works on oscillatory aquatic locomotion since the pioneering work by Lighthill (1970). Particularly, we focus on models where the oscillating foil is elastically connected to the virtual body, simulating the flexibility of the muscles that move laterally or vertically the tail of some aquatic animals, so that a complete or partially passive oscillation of the fin is allowed to generate, as mentioned above, a very noticeable increase in thrust and propulsive efficiency.

The model allows for both passive heave and pitch, in addition to passive flexural deformation of the foil. p. 2
Where the paper says it

Here we consider an analytical VB-OF self-propulsion model with a flexible foil elastically mounted on its leading edge through torsional and translational springs and dampers, thus allowing for both passive heave and pitch, in additional to the passive flexural deformation of the foil, and actuated by a driving oscillating force that simulates the muscle force that generates the lateral (or vertical) tail motion, instead of assuming a prescribed heaving motion in a rigid foil like in the previously mentioned studies. The inclusion of passive heave and deformation allows to expand the parametric ranges to search analytically for the optimal locomotion conditions.

Analytical Formulation

The fluid force and moments are derived by solving the Euler-Bernoulli beam equation along the chord length multiplied by 1 and 1+, using a small displacement approximation. The non-dimensional form of the foil displacement is given by Equation (2. p. 3

Where the paper says it

This approximation covers up to the second bending natural frequency of the fluid-foil system, and it was shown to provide accurately the FSI for values of the non-dimensional stiffness (defined below) from infinity, corresponding to a rigid foil, down to order of 10−1 . The balances of the lateral (or vertical) forces on the foil and of their moments about its leading edge can be obtained by integrating the two-dimensional Euler-Bernoulli beam equation along its chord length multiplied by 1 and 1+, respectively, when the displacement (2.

1). The balances of the lateral forces and moments are obtained by integrating this displacement equation along its chord length multiplied by 1 and 1+, respectively. p. 3
Where the paper says it

This approximation covers up to the second bending natural frequency of the fluid-foil system, and it was shown to provide accurately the FSI for values of the non-dimensional stiffness (defined below) from infinity, corresponding to a rigid foil, down to order of 10−1 . The balances of the lateral (or vertical) forces on the foil and of their moments about its leading edge can be obtained by integrating the two-dimensional Euler-Bernoulli beam equation along its chord length multiplied by 1 and 1+, respectively, when the displacement (2.

Solution Procedure

The system of linear equations for the harmonic motion is written in matrix form A · d0 = b. The solution d0 can be written in amplitude and phase as h0 = h −, 0 = 0, 1 = 1, 2 = 2. p. 5

Where the paper says it

The solution d0 of Eq. 13) can be written, in amplitude and phase, as ℎ0 = ℎ − , 0 = 0 , 10 = 1 1 , 20 = 2 2 .

The reduced frequency is obtained by minimizing det(A). p. 6
Where the paper says it

However, the original system (2. 13), and more precisely the matrix A, is also needed to obtain the non-dimensional natural frequency by minimizing | det(A)|, because in that notation the frequency only enters in the nondimensional parameter .

A procedure involving an iterative process is used to obtain the stride length and all other self-propulsion performance parameters. p. 6
Where the paper says it

This will provide the stride length (2. 11), and all the other self-propulsion performance parameters described in the next section, for the given set of non-dimensional governing parameters.

Performance Analysis

The non-dimensional input power per unit span needed to generate passive motion is computed from Equation (3.1). The ratio of the stride length and the input power coefficient, denoted as ≡, is used as a dimensionless measure of locomotion efficiency. p. 7

Where the paper says it

For that reason, it is preferred here the use of a dimensionless form of the first of these quantities, the ratio of the stride length and the input power coefficient, = . 3) Though dimensionless, this quantity is not an efficiency in the strict sense because it can be larger than unity.

The Strouhal number is defined as the frequency multiplied by the beat amplitude divided by the cruising velocity. p. 7
Where the paper says it

In any case, as we shall see in the reported results, and are always closely related, so that is usually enough to characterize the swimming efficiency. Another relevant non-dimensional parameter that characterizes flapping foil selfpropulsion is the Strouhal number, defined as the frequency multiplied by the beat amplitude (maximum peak-to peak trailing edge amplitude, || = | ( = 1)|, divided by the cruising velocity : St = || = |ℎ0 − 20 + 4810 + 35220| .

Key Findings

The inclusion of passive heave and deformation allows for an expanded parametric range for optimal self-propulsion conditions. A new local optimal locomotion condition is found near the translational spring branch of the elastic support resonance at lower stiffnesses. p. 0

Where the paper says it

The inclusion of passive heave and deformation allows to expand the parametric ranges for optimal self-propulsion conditions in terms of length travelled by flapping cycle (stride length) and locomotion efficiency. In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs.

The best propulsion performance is attained near the translational spring branch for small non-dimensional frequency and small non-dimensional stiffnesses. p. 1
Where the paper says it

(2026) presented a similar VB-OF analytical self-propulsion model to that developed by Zhong et al. (2021) for a rigid foil elastically mounted at its leading edge through a torsional spring, discussing the best performance in terms of the stride length and efficiency as a function of an appropriate non-dimensional frequency, and analyzing the role of fluid damping on the optimal locomotion conditions.

The inclusion of flexibility enhances the maxima of stride length and efficiency associated with the translational spring mode for small frequencies and stiffnesses. p. 0
Where the paper says it

The inclusion of passive heave and deformation allows to expand the parametric ranges for optimal self-propulsion conditions in terms of length travelled by flapping cycle (stride length) and locomotion efficiency. In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs.

Conclusion

The present analytical model is a useful tool to characterize the best self-propulsion performance for any set of given parameters, provided that the resulting pitch, heave and deformation amplitudes remain small. The inclusion of passive heave and deformation increases the possibility of a better locomotion performance by enlarging the parameters space. p. 0

Where the paper says it

The inclusion of passive heave and deformation allows to expand the parametric ranges for optimal self-propulsion conditions in terms of length travelled by flapping cycle (stride length) and locomotion efficiency. In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs.

A new optimal parametric region for even better propulsion performance exists around the translational spring branch. p. 0
Where the paper says it

The inclusion of passive heave and deformation allows to expand the parametric ranges for optimal self-propulsion conditions in terms of length travelled by flapping cycle (stride length) and locomotion efficiency. In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs.

The best self-propulsion performance is attained by a not too rigid foil for small values of non-dimensional frequency and stiffness located around the translational spring branch. p. 1
Where the paper says it

(2026) presented a similar VB-OF analytical self-propulsion model to that developed by Zhong et al. (2021) for a rigid foil elastically mounted at its leading edge through a torsional spring, discussing the best performance in terms of the stride length and efficiency as a function of an appropriate non-dimensional frequency, and analyzing the role of fluid damping on the optimal locomotion conditions.

The approximation (4. p. 3
Where the paper says it

1) where ℎ() and () characterize the (unknowns) heaving and pitching motions of the pivot axis located at the leading edge ( = −1), and 1 () and 2 () are two additional unknown amplitudes characterizing the passive flexural deformation of a foil with a free trailing edge, ( 2 /2 )=1 =( 3 /3 )=1 = 0. This approximation covers up to the second bending natural frequency of the fluid-foil system, and it was shown to provide accurately the FSI for values of the non-dimensional stiffness (defined below) from infinity, corresponding to a rigid foil, down to order of 10−1 .

7) for the springs natural frequencies remains a good approximation as stiffness decreases within the limits of the present model. p. 0
Where the paper says it

In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs. Unlike the local maximum of efficiency close to the natural frequency associated with the torsional spring branch, which increases with the stiffness of the foil, being the highest for a rigid foil, the larger local maximum associated with the translational spring branch increases as the stiffness of the foil decreases.

The best self-propulsion performance is attained near the springs resonance for small non-dimensional frequency and small non-dimensional stiffnesses. p. 1
Where the paper says it

(2026) presented a similar VB-OF analytical self-propulsion model to that developed by Zhong et al. (2021) for a rigid foil elastically mounted at its leading edge through a torsional spring, discussing the best performance in terms of the stride length and efficiency as a function of an appropriate non-dimensional frequency, and analyzing the role of fluid damping on the optimal locomotion conditions.

The highest locomotion efficiency is reached near the translational spring branch for small non-dimensional frequency and stiffnesses. p. 0
Where the paper says it

In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs. Unlike the local maximum of efficiency close to the natural frequency associated with the torsional spring branch, which increases with the stiffness of the foil, being the highest for a rigid foil, the larger local maximum associated with the translational spring branch increases as the stiffness of the foil decreases.

The best self-propulsion performance found with the present model is thus attained by a not too rigid foil for small values of non-dimensional frequency and stiffness located around the translational spring branch of the springs resonant frequency. p. 0
Where the paper says it

In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs. Unlike the local maximum of efficiency close to the natural frequency associated with the torsional spring branch, which increases with the stiffness of the foil, being the highest for a rigid foil, the larger local maximum associated with the translational spring branch increases as the stiffness of the foil decreases.

The inclusion of passive heave and deformation allows to expand the parametric ranges to search analytically for the optimal locomotion conditions. p. 2
Where the paper says it

Here we consider an analytical VB-OF self-propulsion model with a flexible foil elastically mounted on its leading edge through torsional and translational springs and dampers, thus allowing for both passive heave and pitch, in additional to the passive flexural deformation of the foil, and actuated by a driving oscillating force that simulates the muscle force that generates the lateral (or vertical) tail motion, instead of assuming a prescribed heaving motion in a rigid foil like in the previously mentioned studies. The inclusion of passive heave and deformation allows to expand the parametric ranges to search analytically for the optimal locomotion conditions.

The best propulsion performance with more flexibility is attained associated to the translational branch of the springs natural modes. p. 0
Where the paper says it

In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs. Unlike the local maximum of efficiency close to the natural frequency associated with the torsional spring branch, which increases with the stiffness of the foil, being the highest for a rigid foil, the larger local maximum associated with the translational spring branch increases as the stiffness of the foil decreases.

The deformation of the flexible foil allows to get a given thrust with smaller pitch and heave amplitudes than for a rigid foil. p. 2
Where the paper says it

Here we consider an analytical VB-OF self-propulsion model with a flexible foil elastically mounted on its leading edge through torsional and translational springs and dampers, thus allowing for both passive heave and pitch, in additional to the passive flexural deformation of the foil, and actuated by a driving oscillating force that simulates the muscle force that generates the lateral (or vertical) tail motion, instead of assuming a prescribed heaving motion in a rigid foil like in the previously mentioned studies. The inclusion of passive heave and deformation allows to expand the parametric ranges to search analytically for the optimal locomotion conditions.

The present analytical model is an excellent approximation to analyze the locomotion performance of this kind of fish-like robots, even for values of the heave amplitude as high as h0 = 1. p. 1
Where the paper says it

(2021) developed a VB-OF self-propulsion model to explain their experimental results using a tuna-like platform that tunes its own tail stiffness using a motor-driven ‘muscle’. Their experiments showed that to maximize efficiency, muscle tension should scale with swimming speed squared, offering a simple tuning strategy for this kind of fish-like robots.

28. The present model has the advantage, in relation to these previous ones, – independently of, and in addition to, a presumably better accuracy of the present model for a wider range. p. 9
Where the paper says it

The stride length from the present model for (continuous line) is practically the same for the maximum cruising velocity, that occurs near the resonance frequency of the torsional spring, but differs slightly as the frequency departs from it. It must be noted that the present for a rigid foil has shown a better general agreement with numerical and experimental results than Garrick’s model for the timeaveraged thrust coefficient (e.

How the training signal is built

The non-dimensional parameters are redefined by marking them with a hat ‘ˆ’ to simplify the system of linear equations. The new, hatted inertial matrix Aˆ 0 is independent of ≡, which simplifies the system. p. 6

Where the paper says it

20) (and, of course, in ) through the non-dimensional parameter . Actually, the new, hatted inertial matrix Aˆ 0 is independent of (see Eq.

The computing procedure involves starting with a low value of ≡ and solving the system to obtain the unknowns. p. 6
Where the paper says it

22) In the present self-propulsion problem, this ratio is also part of the solution, but sometimes it is convenient to know how the other unknowns are related to the resonant frequency of the system rather than in terms of the input parameters. The computing procedure for a given set of the governing non-dimensional parameters, that is, for given , ˆ, ˆℎ, ˆ ℎ, ˆ, ˆ and , can be summarized as follows: (i) Start assuming a (low) value of and solve (2.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Incorporate a solver for Eq (2.13) that iteratively adjusts non-dimensional parameters to satisfy Eq (2.14), allowing it to find stride length and cruising velocity based on given governing parameters.

  2. Develop a system that uses the analytical approximation of the resonant frequency given by Eq (4.7) to identify two local maxima in propulsion performance corresponding to the translational spring branch and torsional spring branch for a rigid foil.

  3. Enhance the system's capability to predict optimal locomotion conditions by analyzing Figure 9, which shows how stride length and efficiency increase with stiffness for different modes (H) and (P).

  4. Implement a comparison module that contrasts the performance of the current model against previous work, such as Zhong et al. (2021), showing where the current model provides better accuracy for a wider range of input force amplitudes and stiffnesses.

  5. Integrate a mechanism to estimate the non-dimensional natural frequency by minimizing det(A), which is useful for representing results in terms of the ratio between actuating frequency and natural frequency.

Abstract

An analytical formulation of the fluid-structure interaction of a flexible foil driven by an oscillating force actuating on its elastically mounted leading edge, so that it can heave, pitch and deform passively with the hydrodynamic forces, is used to investigate the aquatic locomotion of a body, responsible for the whole drag and thrusted by the oscillating flexible foil. The small-amplitude theoretical model is validated with previous theoretical and experimental results for a body propelled by a rigid plate oscillating with a prescribed heaving motion and passive pitch. The inclusion of passive heave and deformation allows to expand the parametric ranges for optimal self-propulsion conditions in terms of length travelled by flapping cycle (stride length) and locomotion efficiency. In addition to the known optimal locomotion condition localized near the resonance of the torsional spring on which the foil is elastically mounted, which here is modulated by its coupling with the resonances of the translational spring and of the structural deformation of the foil, another even better local optimal locomotion condition is found near the translational spring branch of the elastic support resonance that occurs at lower stiffnesses of both springs. Unlike the local maximum of efficiency close to the natural frequency associated with the torsional spring branch, which increases with the stiffness of the foil, being the highest for a rigid foil, the larger local maximum associated with the translational spring branch increases as the stiffness of the foil decreases.

Related papers