Muscular Hydrostat Models

• Muscular hydrostat models are biological structures that use intricate muscle fiber architectures and volume preservation to enable large, controlled deformations without skeletal support.
• They employ continuum mechanics approaches, such as Cosserat rod and thin shell theories, to couple axial, radial, and bending strains under isovolumetric constraints.
• Validated in systems like octopus arms and soft robotic analogs, these models inform the design of actuators with tunable stiffness and effective shape control.

A muscular hydrostat is a biological structure, such as an octopus arm, elephant trunk, Drosophila larva, or mammalian tongue, that relies on an intricate architecture of muscle fibers to achieve controlled, large-amplitude deformations in the absence of rigid skeletal support. The core mechanical principle is that global incompressibility of the tissue (constant volume constraint) couples muscle-driven axial, radial, and bending strains, yielding both variable stiffness and actuator redundancy. Recent advances in continuum mechanics, multi-field control, and mathematical modeling have enabled the construction of detailed, validated models that reveal how complex behaviors—ranging from rolling and bend propagation to grasping and object transport—are generated and controlled in both biological and bio-inspired soft-bodied systems.

1. Geometric and Mechanical Foundations

Muscular hydrostat models universally adopt a continuum mechanics framework, typically leveraging Cosserat rod or thin shell/membrane theory to describe the body plan. For instance, the Drosophila larva is idealized as a thin-walled cylinder of length $L_0$ and radius $R_0$, discretized into $N$ segments, with local cross-sectional area $A(s)=\pi R(s)^2$ and wall thickness $h\ll R$ (Liang et al., 2024). The configuration is parameterized by midline curvature $\kappa(s,t)$ or discrete segment angles $\theta_i(t)$, where $\kappa\approx\partial\theta/\partial s$. Bending kinematics, shear, and extension are similarly treated in models of octopus arms, where the Cosserat rod formalism describes the centerline $r(s,t)$, local material frame $R(s,t)\in$ SO(3), and strain measures for extension $\varepsilon(s,t)$, bending $(\kappa_1,\kappa_2)$, and twist $\tau(s,t)$ (Tekinalp et al., 2023, Sun et al., 2024).

Extended Cosserat models incorporate a cross-sectional inflation ratio $\rho(s,t)$, representing in-plane expansion or contraction and thereby accounting for lateral deformation modes critical in hydrostatic function (Sun et al., 2024, Golestaneh et al., 2024).

2. Volume Preservation and Hydrostatic Constraints

The defining feature of a muscular hydrostat is the isovolumetric constraint, which mathematically enforces conservation of volume, despite large shape deformations. This is stated, for example, as $V=\int_0^{L} A(s,t)\,ds = \text{const}$, or on differential segments by $(1+\epsilon_a)(1+\epsilon_r)^2=1$, yielding the familiar approximation $\epsilon_r\approx-\frac{1}{2}\epsilon_a$ for small strains (Liang et al., 2024). In the full 3D Cosserat-rod setting, incompressibility is imposed by requiring $\det F=1$, where $F$ is the deformation gradient (Tekinalp et al., 2023, Sun et al., 2024). This constraint couples axial strain to radial (or cross-sectional) strain, so muscle-driven shortening in one direction necessitates compensatory expansion in another.

In numerical models of octopus-inspired soft arms under water, volume preservation is enforced via an ODE for the cross-sectional stretch $\beta(s)$, entering into both the strain energy and the stress resultants, which directly modulate the arm's stiffness (Golestaneh et al., 2024). This constraint is fundamental for tuning mechanical response, facilitating stiffness modulation, and setting kinematic bounds on curvature, elongation, and force transmission (Liang et al., 2024, Sun et al., 2024).

3. Muscle Architecture, Activation, and Constitutive Laws

Hydrostatic operation relies on both the arrangement and activation of muscle fibers. Anatomical realism is achieved by modeling multiple muscle groups: longitudinal, transverse (circumferential), and, in three-dimensional models, oblique (helical) families (Tekinalp et al., 2023, Chang et al., 2020). Each group contributes uniquely—longitudinal fibers drive length changes and bending; transverse muscles regulate cross-sectional area, and oblique fibers induce twist.

Muscle actuation enters as a neural activation field $\alpha(s,t)\in [0,1]$ (or $a^m(s,t)$ for group $m$). Activated muscle fibers shorten by $$\Delta L_m(s,t) = \alpha(s,t)\epsilon_{m,\max}L_{m,0}$$, resulting in axial strains and corresponding contractile forces, typically modeled using Hill-type or force–length curves (Liang et al., 2024, Chang et al., 2020). The overall stress is a combination of passive elastic response (e.g., $$W(\epsilon,\kappa,\tau)=\frac{1}{2}E\,A\,\epsilon^2+EI(\kappa_1^2+\kappa_2^2)+GJ \tau^2$$) and active, muscle-generated stress (Tekinalp et al., 2023). Internal hydrostatic pressure $p(s,t)$ is coupled to muscle contraction via membrane stress balance, e.g., $p(s)=Eh/R(\nu-\frac{1}{2})\epsilon_a(s)$ (Liang et al., 2024).

Constitutive laws often include hyperelastic (e.g., Saint-Venant Kirchhoff) and viscoelastic (Kelvin–Voigt) terms for biological fidelity, especially in dynamic modeling and when lateral deformation is non-negligible (Sun et al., 2024).

4. Dynamic Equations and Environmental Coupling

The core mechanical evolution is expressed as coupled PDEs (or reduced ODEs) representing conservation of linear and angular momentum, with active and passive forces, moments, and pressures, as well as environmental loads (friction, drag, weight, buoyancy). For soft-body rolling, translational and rotational equations capture center-of-mass motion and rolling induced by internal muscle moments, e.g.,

$$m\ddot X = F_{\rm drive} - F_{\rm friction}, \quad I\ddot\phi = M_{\rm muscle}(\kappa,\alpha) + M_{\rm env}$$

where $M_{\rm muscle}=\int_0^L [r(s)\times F_m(s)]ds$ (Liang et al., 2024). Fluid-structure interaction is incorporated in aquatic systems, adding terms for hydrodynamic drag and added mass (Golestaneh et al., 2024, Sun et al., 2024).

Passive compliance and environmental forces are not mere perturbations; they are integral to self-stabilization, robustness, and the capacity for shape adaptation, especially during contact-rich tasks such as grasping or crevice probing (Tekinalp et al., 2023).

5. Control and Activation Waveforms

Efficient actuation of muscular hydrostats is accomplished via spatiotemporal activation patterns that exploit the body's mechanical redundancy. Traveling-wave activation of muscle groups—whether as a circumferential wave in rolling (Liang et al., 2024), a base-to-tip bend in octopus arms (Tekinalp et al., 2023, Chang et al., 2020, Wang et al., 2021), or localized pulses—allows generation and transport of curvature, twist, and extension with a small number of low-dimensional control signals.

In octopus arms, ~200 muscle groups can be synchronized or independently triggered through "activation templates": uniform contraction (length/tension), traveling wavefronts (bending), or localized pulses (twist), and synthesized by hand-tuned amplitude, duration, and wave speed (Tekinalp et al., 2023). In the context of energetic shape control, the activation field $\alpha(s,t)$ becomes the control variable in a Hamiltonian framework, and optimal control can be reduced to bilevel optimization: an inner level solving for equilibrium shapes, and an outer level optimizing activation profiles for task objectives (Chang et al., 2020).

Reduced-order models use parametric ansatzes for curvature, e.g., $\kappa(s,t)=\phi(s;\mu(t),\epsilon(t))$, dramatically lowering control dimensionality while retaining biomechanical realism, as validated against high-fidelity simulations and experimental data (Wang et al., 2021).

6. Topological and Task-Level Descriptions

Controlling a muscular hydrostat’s configuration in high-dimensional shape space is challenging. Topological invariants—such as linking number $L$, writhe $W$, and twist $T$ of the centerline and auxiliary curves—permit a task-level abstraction. The Călugăreanu–White–Fuller theorem $L=W+T$ applies directly to octopus arms, where shape programming becomes a matter of steering $W$ and $T$ via selected muscle groups (Tekinalp et al., 2023). These quantities encode object grasping (e.g., coil number), in-hand manipulation, and the avoidance of self-intersection during complex fetch or probe maneuvers.

This topological perspective allows for a drastic reduction in control dimensionality, as transitions between shape classes (helix $\leftrightarrow$ spiral) are mapped onto controlled transport between $W$ and $T$ without recourse to full trajectory planning.

7. Applications, Validation, and Extensions

Canonical muscular hydrostat models have been experimentally validated in biological and soft-robotic systems. For example, Drosophila larval rolling was replicated using pneumatic soft robots, confirming the sufficiency of sequential neural activation and hydrostatic skeleton mechanics for efficient rolling without external torque (Liang et al., 2024). In underwater octopus-arm analogs, twisted and coiled artificial muscles actuate a continuum Cosserat rod with drag, producing realistic bend magnitudes and kinematics in agreement with finite-element simulations (Golestaneh et al., 2024).

Stiffness tuning via inflation (adjusting $\rho(s,t)$ in the extended Cosserat model) enables variable compliance, critical for manipulation tasks and demonstrated in both simulated and robotic contexts (Sun et al., 2024). The modeling strategies readily transfer to elephant trunks and mammalian tongues, with only modification of muscle layout, environmental loads, and relevant geometric constraints (Wang et al., 2021).

Across all domains, key generalizations emerge:

• Isovolumetric architecture guarantees kinematic coupling of extension, contraction, and bending.
• Internal pressure offers tunable, spatially heterogeneous stiffness.
• Sequential activation drives efficient, self-sustained locomotion and manipulation.
• Topological control supersedes high-dimensional trajectory planning, enabling scalable strategies for practical soft-robotic systems (Tekinalp et al., 2023, Sun et al., 2024).