Cetaev Condition in Nonholonomic Mechanics

• Cetaev condition is a criterion that defines admissible virtual displacements by requiring the restricted variation δ^(c)g=0 in nonholonomic systems.
• It uses the restricted variation instead of the full variational formulation to derive d’Alembert–Lagrange equations across nonlinear, linear, and homogeneous constraints.
• The condition enables the formulation of motion equations with constraint forces and clarifies differences with vakonomic approaches in complex mechanical systems.

The Cetaev condition is a foundational construct in the analysis of nonholonomic mechanical systems, providing a criterion for admissible virtual displacements in the presence of nonintegrable velocity constraints. It establishes a prescription—distinct from that used for holonomic systems—for the class of virtual displacements upon which the d’Alembert–Lagrange equations of motion are built. The relevance of the Cetaev condition spans nonlinear, homogeneous, and linear nonholonomic constraints, with critical implications for the derivation and interpretation of physically consistent equations of motion in variational and non-variational frameworks (Talamucci, 30 Dec 2025, Talamucci, 11 Jun 2025, Talamucci, 2024, Talamucci, 2024, Talamucci, 2023).

1. Mathematical Statement and Definition

For a mechanical system with generalized coordinates $q=(q_1,\dots,q_n)$ and $\kappa < n$ independent velocity constraints,

$$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$

where

$\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$

the Cetaev condition asserts that the allowed virtual displacements $\delta q = (\delta q_1, \dots, \delta q_n)$ must satisfy

$$\boxed{ \delta^{(c)} g_\nu = \sum_{i=1}^n \frac{\partial g_\nu}{\partial \dot{q}_i} \, \delta q_i = 0, \quad \nu=1, \dots, \kappa }$$

This is formulated at the level of infinitesimal variations rather than as a consequence of the constraint equations themselves (Talamucci, 30 Dec 2025, Talamucci, 11 Jun 2025).

Two types of variations are involved: $$\delta^{(c)} F := \sum_{i=1}^n \frac{\partial F}{\partial \dot{q}_i} \, \delta q_i , \qquad \delta^{(v)} F := \sum_{i=1}^n \frac{\partial F}{\partial q_i} \, \delta q_i + \sum_{i=1}^n \frac{\partial F}{\partial \dot{q}_i} \, \delta \dot{q}_i.$$

2. Variational Basis and Transpositional Rule

The Cetaev condition is grounded in the requirement that the virtual work of the constraint forces vanishes on the subspace defined by $\delta^{(c)} g_\nu = 0$. Unlike vakonomic (variational) formulations, which employ the full variation $\delta^{(v)}g_\nu = 0$, the Cetaev prescription only imposes the restricted condition $\delta^{(c)}g_\nu = 0$, with no a priori assumption of a commutation relation between $\kappa < n$0 and $\kappa < n$1.

A fundamental identity, the transpositional (or commutation) rule, relates $\kappa < n$2 and $\kappa < n$3: $\kappa < n$4 where

$\kappa < n$5

is the $\kappa < n$6-th Lagrangian derivative (Talamucci, 30 Dec 2025, Talamucci, 11 Jun 2025). Enforcing the Cetaev rule yields the orthogonality of admissible virtual displacements to the rows of $\kappa < n$7, and does not require $\kappa < n$8, unless specifically imposed to align with vakonomic approaches (Talamucci, 2024).

3. Physical Interpretation and Special Cases

The Cetaev condition generalizes the virtual displacement prescription of holonomic systems. For holonomic constraints $\kappa < n$9, the Cetaev rule reduces to the traditional condition $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$0. For linear constraints,

$$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$1

Cetaev’s condition simplifies to $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$2, identifying $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$3-dimensional subspaces of admissible $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$4 (Talamucci, 30 Dec 2025, Talamucci, 11 Jun 2025, Talamucci, 2023).

For constraints homogeneous in velocities, i.e., $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$5, Euler’s theorem ensures that $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$6 on the constraint manifold. The Cetaev prescription then gives a rule compatible with the energy conservation and interpretation familiar from holonomic theory (Talamucci, 2023).

4. Consistency, Commutation, and Vakonomic Comparison

Consistency between the Cetaev condition and full variational principles hinges upon the properties of the Lagrangian derivatives $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$7. In general, only for integrable, linear, or homogeneous constraints can simultaneous enforcement of the conditions

$$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$8

and $$g_\nu(q, \dot{q}, t) = 0 , \qquad \nu=1, \dots, \kappa,$$9 ensure the commutation $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$0, a requirement for equivalence with vakonomic or extended variational formulations (Talamucci, 30 Dec 2025, Talamucci, 2024). For genuinely nonlinear, nonintegrable constraints, the Cetaev and vakonomic equations diverge, and any derivation of the Cetaev rule from algebraic manipulations of the constraints is only rigorously justified in special cases (e.g., homogeneous or integrable constraints), not in general (Talamucci, 2024, Talamucci, 2024).

5. Equations of Motion and Algebraic Framework

Applying the Cetaev condition to the d’Alembert–Lagrange principle,

$\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$1

for all $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$2 in the Cetaev subspace, one obtains the equations: $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$3 with $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$4 as undetermined multipliers. This form is immediate for linear and homogeneous constraints. For systems with nonlinear constraints, elimination of dependent velocities and application of Cetaev’s rule yields reduced Lagrangians and corresponding motion equations that retain the structure of ideal constraint forces (Talamucci, 2023, Talamucci, 2024).

6. Domain of Validity, Limitations, and Examples

The Cetaev condition supplies a minimal, physically motivated definition for admissible variations in nonholonomic systems but constitutes an independent axiom rather than a theorem provable from the constraint equations except under integrability or homogeneity (Talamucci, 2024). Where $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$5, commutation of $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$6 and $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$7 cannot, in general, be imposed without additional structure. Consequently, for generic nonlinear velocity constraints, equations of motion derived from d’Alembert–Cetaev and vakonomic methods may differ unless an integrating factor or other special circumstances apply (Talamucci, 30 Dec 2025, Talamucci, 11 Jun 2025).

Illustrative examples include:

• Linear constraints: $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$8.
• Homogeneous quadratic constraints: $\rank\left(\frac{\partial g_\nu}{\partial \dot{q}_i}\right) = \kappa,$9.
• Nonholonomic constraints in multi-particle systems that are homogeneous in velocities, exhibiting physically coherent energy integrals and virtual work properties (Talamucci, 2023).

7. Summary and Theoretical Implications

The Cetaev condition remains indispensable for rigorously formulating the equations of motion for nonholonomic systems within the d’Alembert–Lagrange framework, especially in contexts involving nonlinear and homogeneous velocity constraints. Its status is fundamentally axiomatic except when additional mathematical structure—integrability, linearity, or suitable homogeneity—permits its derivation or provides deeper geometric meaning. The ongoing delineation of its limitations and the explicit quantification of its relation to variational (vakonomic) methods underscore its centrality, while simultaneously clarifying the boundaries of its applicability in advanced theoretical and applied nonholonomic dynamics (Talamucci, 30 Dec 2025, Talamucci, 11 Jun 2025, Talamucci, 2024, Talamucci, 2024, Talamucci, 2023).