Inverse problem of the calculus of variations on Lie groups (Q2580932)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Inverse problem of the calculus of variations on Lie groups |
scientific article |
Statements
Inverse problem of the calculus of variations on Lie groups (English)
0 references
10 January 2006
0 references
Let \(G\) denote a connected Lie group, and let \(\mathfrak{G}\) denote its Lie algebra of left invariant vector fields on \(G\). \(G\) admits a canonical connection \(\nabla\) defined by \(\nabla_{X}Y = \frac{1}{2}[X,Y]\) for all \(X,Y \in \mathfrak{G}\). The connection \(\nabla\) is bi-invariant and has a parallel curvature tensor \(R\) given by \(R(X,Y)Z = \frac{1}{4}[Z,[X,Y]]\) for all \(X,Y,Z \in \mathfrak{G}\). In particular \(R \equiv 0 \Leftrightarrow\mathfrak{G}\) is 2-step nilpotent. For a \(C^{\infty}\) manifold \(M\), a function \(E : TM \rightarrow \mathbb{R}\) is a {Lagrangian} if \(E\) is smooth on \(TM\) and det(\(\partial ^{2} E/\partial y^{\alpha}y^{\beta}) \neq 0\). Regular Lagrangians determine vector fields \(S\) on \(TM\) called sprays whose integral curves are solutions to the Euler-Lagrange equations that arise from \(E\). In this article, the authors consider the following question: On a connected Lie group \(G\) can the velocity fields of the geodesics of the canonical connection \(\nabla\) be realized as the integral curves of the spray \(S\) associated to some regular smooth Lagrangian \(E : TG \rightarrow \mathbb{R}\)? The authors obtain an affirmative answer in the following cases: 1) \(G\) is the group of rigid motions of the plane \(\mathbb{R}^{2}\). Two examples of the desired Lagrangians are given. 2) \(G\) is a 2-step nilpotent Lie group. Furthermore, an explicit example of a desired Lagrangian is given if \(G\) is a Heisenberg algebra (odd-dimensional with 1-dimensional center). The authors obtain a negative answer to the question if \(G\) is the 6-dimensional group of affine transformations of the plane \(\mathbb{R}^{2}\). In general the answer to the question is very difficult to determine. The authors address the question with methods that involve the Froehlicher-Nijenhuis theory of derivations of the exterior algebra of differential forms and the Spencer-Goldschmidt theory of overdetermined systems of partial differential equations.
0 references
Lie group
0 references
canonical connection
0 references
Euler-Lagrange equation
0 references
variational principle
0 references