Variational Principles for Hamiltonian Systems

Brian K. Tran, Melvin Leok

Abstract

Motivated by recent developments in Hamiltonian variational principles, Hamiltonian variational integrators, and their applications such as to optimization and control, we present a new Type II variational approach for Hamiltonian systems, based on a virtual work principle that enforces the Type II boundary conditions through a combination of essential and natural boundary conditions; particularly, this approach allows us to define this variational principle intrinsically on manifolds. We first develop this variational principle on vector spaces and subsequently extend it to parallelizable manifolds, general manifolds, as well as to the infinite-dimensional setting. Furthermore, we provide a review of variational principles for Hamiltonian systems in various settings as well as their applications.

Introduction

In classical mechanics, the equations of motion are typically derived using Hamilton’s Principle, which involves finding a path that makes the "action" functional stationary given fixed starting and ending positions (Type I boundary conditions). While this framework is foundational, it encounters significant theoretical and practical hurdles when applied to modern problems in optimization, control, and degenerate physical systems.

This paper introduces a robust "Type II" variational framework designed to operate intrinsically on curved manifolds. By shifting the focus from stationary action to a virtual work principle, the authors provide a mathematical foundation that accommodates systems where traditional methods fail, particularly those with degenerate Hamiltonians or complex geometric constraints.

The Problem: Degeneracy and Manifolds

The standard Type I approach (fixing $q(0)$ and $q(T)$) is often insufficient for two primary reasons:

  1. Degenerate Systems: Many important systems, such as those used in adjoint sensitivity analysis or certain constrained dynamics, have "degenerate" Hamiltonians. In these cases, fixing both the start and end positions is mathematically inconsistent—the system becomes simultaneously overdetermined and underdetermined.
  2. Geometric Constraints on Manifolds: Defining variational principles on curved spaces (manifolds) is notoriously difficult. In a Type II approach, one typically fixes the initial position $q(0)$ and the final momentum $p(T)$. However, on a manifold, a momentum vector is an element of the cotangent space $T^*_{q(T)}Q$, which is tied to the specific base point $q(T)$. If the final position $q(T)$ is unknown or variable, defining a "fixed" final momentum becomes difficult without relying on specific coordinate charts.

Methodology: The Virtual Work Principle

To overcome these issues, the authors move away from the standard "stationary action" approach and instead adopt a Type II d’Alembert Variational Principle based on virtual work.

Instead of requiring the action functional itself to be stationary, they require the variation of the action to equal the "virtual work" done by the boundary momentum. This formulation allows the boundary conditions to emerge as "natural" consequences of the variation rather than "essential" constraints imposed at the start.

Extension to Manifolds

For parallelizable manifolds (such as Lie groups), the authors derive the Hamel equations, which are the non-canonical equivalents of Hamilton's equations. For more general manifolds that are not parallelizable, they treat the system as a "free boundary problem." In this setting, the final momentum is defined as a section of the cotangent bundle (a field of possible momenta), allowing the final position to remain variable while still maintaining a well-defined, coordinate-independent variational principle.

Key Insights: Completeness and Adjoint Sensitivity

A critical contribution of this work is the analysis of "completeness" for different boundary condition types in the context of degenerate systems.

Table 1: Classification of boundary conditions and their completeness for degenerate Hamiltonian systems. Types II and III are shown to be complete, justifying their use over standard Type I principles in degenerate scenarios.

As illustrated in the table above, the authors prove that for a model degenerate Hamiltonian, Type I and Type IV boundary conditions are fundamentally incompatible with the physics (labeled "Incomplete"), whereas Type II and III are "Complete." This justifies the shift toward Type II formalisms for complex systems.

Furthermore, the paper highlights that "maximally degenerate" Hamiltonians are exactly the forms used in adjoint sensitivity analysis. This means their new variational principle provides the natural geometric setting for understanding how sensitive a system's output is to its initial conditions, linking classical mechanics directly to modern control theory.

Conclusion and Future Work

The authors have successfully derived a Type II variational principle that transitions seamlessly from simple vector spaces to parallelizable manifolds, general manifolds, and finally to infinite-dimensional reflexive Banach spaces. By demonstrating that the free boundary variational principle becomes a standard stationarity condition when the boundary momentum is chosen as the derivative of a cost function ($p_1 = dC$), they provide a unified framework for Hamiltonian dynamics and optimal control.

Future research directions include:

  • Numerical Implementation: Building discrete-time "Type II Variational Integrators" that preserve the symplectic structure of Hamiltonian systems on curved spaces.
  • Stochastic Systems: Extending these intrinsic manifold-based principles to stochastic Hamiltonian systems to account for uncertainty in complex dynamics.

Summary updated

Abstract

Motivated by recent developments in Hamiltonian variational principles, Hamiltonian variational integrators, and their applications such as to optimization and control, we present a new Type II variational approach for Hamiltonian systems, based on a virtual work principle that enforces the Type II boundary conditions through a combination of essential and natural boundary conditions; particularly, this approach allows us to define this variational principle intrinsically on manifolds. We first develop this variational principle on vector spaces and subsequently extend it to parallelizable manifolds, general manifolds, as well as to the infinite-dimensional setting. Furthermore, we provide a review of variational principles for Hamiltonian systems in various settings as well as their applications.
BibTeX
@inproceedings{tran2025variational,
  author = {Brian K. Tran and Melvin Leok},
  title = {{Variational Principles for Hamiltonian Systems}},
  booktitle = {Geometric},
  year = {2025},
  doi = {10.1142/s2972458925500042},
}
Figure