Finsler manifold

In mathematics, particularly differential geometry, a Finsler manifold is a differentiable manifold together with the structure of an intrinsic quasimetric space in which the length of any rectifiable curve γ : [a,b] → M is given by the length functional

L[\gamma] = \int_a^b F(\gamma(t),\dot{\gamma}(t))\,dt,

where F(x, · ) is a Minkowski norm (or at least an asymmetric norm) on each tangent space TxM. Finsler manifolds non-trivially generalize Riemannian manifolds in the sense that they are not necessarily infinitesimally Euclidean. This means that the (asymmetric) norm on each tangent space is not necessarily induced by an inner product (metric tensor).

Élie Cartan (1933) named Finsler manifolds after Paul Finsler, who studied this geometry in his dissertation (Finsler 1918).

Contents

Definition

A Finsler manifold is a differentiable manifold M together with a Finsler function F defined on the tangent bundle of M so that for all tangent vectors v,

In other words, F is an asymmetric norm on each tangent space. Typically one replaces the subadditivity with the following strong convexity condition:

Here the hessian of F2 at v is the symmetric bilinear form

\mathbf{g}_v(X,Y)�:= \frac{1}{2}\left.\frac{\partial^2}{\partial s\partial t}\left[F(v %2B sX %2B tY)^2\right]\right|_{s=t=0},

also known as the fundamental tensor of F at v. Strong convexity of F2 implies the subadditivity with a strict inequality if u/F(u) ≠ v/F(v). If F2 is strongly convex, then F is a Minkowski norm on each tangent space.

A Finsler metric is reversible if, in addition,

A reversible Finsler metric defines a norm (in the usual sense) on each tangent space.

Examples

Randers manifolds

Let (M,a) be a Riemannian manifold and b a differential one-form on M with

 \|b\|_a�:= \sqrt{a^{ij}b_i b_j} < 1,

where (a^{ij}) is the inverse matrix of (a_{ij}) and the Einstein notation is used. Then

 F(x,v)�:= \sqrt{a_{ij}(x)v^iv^j} %2B b_i(x)v^i

defines a Randers metric[1] on M and (M,F) is a Randers manifold, a special case of a non-reversible Finsler manifold.

Smooth quasimetric spaces

Let (M,d) be a quasimetric so that M is also a differentiable manifold and d is compatible with the differential structure of M in the following sense:

 \frac{1}{C}\|\varphi(y)-\varphi(x)\| \leq d(x,y) \leq C\|\varphi(y)-\varphi(x)\|.

Then one can define a Finsler function F : TM →[0,∞[ by

F(x,v)�:= \lim_{t\to 0%2B} \frac{d(\gamma(0),\gamma(t))}{t},

where γ is any curve in M with γ(0) = x and γ'(0) = v. The Finsler function F obtained in this way restricts to an asymmetric (typically non-Minkowski) norm on each tangent space of M. The induced intrinsic metric dLM × M → [0, ∞] of the original quasimetric can be recovered from

 d_L(x,y)�:= \inf\left\{\  \left.\int_0^1 F(\gamma(t),\dot\gamma(t)) \, dt \ \right| \ \gamma\in C^1([0,1],M) \ , \ \gamma(0)=x \ , \ \gamma(1)=y \ \right\},

and in fact any Finsler function F : TM → [0, ∞) defines an intrinsic quasimetric dL on M by this formula.

Geodesics

Due to the homogeneity of F the length

L[\gamma]:=\int_a^b F(\gamma(t),\dot{\gamma}(t))\, dt

of a differentiable curve γ:[a,b]→M in M is invariant under positively oriented reparametrizations. A constant speed curve γ is a geodesic of a Finsler manifold if its short enough segments γ|[c,d] are length-minimizing in M from γ(c) to γ(d). Equivalently, γ is a geodesic if it is stationary for the energy functional

E[\gamma]:=\frac{1}{2}\int_a^b F^2(\gamma(t),\dot{\gamma}(t))\, dt

in the sense that its functional derivative vanishes among differentiable curves γ:[a,b]→M with fixed endpoints γ(a)=x and γ(b)=y.

Canonical spray structure on a Finsler manifold

The Euler-Lagrange equation for the energy functional E[γ] reads in the local coordinates (x1,...,xn,v1,...,vn) of TM as

g_{ik}\Big(\gamma(t),\dot\gamma(t)\Big)\ddot\gamma^i(t) %2B \left(\frac{\partial g_{ik}}{\partial x^j}\Big(\gamma(t),\dot\gamma(t)\Big)
- \frac{1}{2}\frac{\partial g_{ij}}{\partial x^k}\Big(\gamma(t),\dot\gamma(t)\Big) \right)\dot\gamma^i(t)\dot\gamma^j(t) = 0,

where k=1,...,n and gij is the coordinate representation of the fundamental tensor, defined as


g_{ij}(x,v)�:= g_v\left(\tfrac{\partial}{\partial x^i}\big|_x,\tfrac{\partial}{\partial x^j}\big|_x\right).

Assuming the strong convexity of F2(x,v) with respect to vTxM, the matrix gij(x,v) is invertible and its inverse is denoted by gij(x,v). Then γ:[a,b]→M is a geodesic of (M,F) if and only if its tangent curve γ':[a,b]→TM \0 is an integral curve of the smooth vector field H on TM \0 locally defined by


H|_{(x,v)}�:= v^i\tfrac{\partial}{\partial x^i}\big|_{(x,v)} - \ 2G^i(x,v)\tfrac{\partial}{\partial v^i}\big|_{(x,v)},

where the local spray coefficients Gi are given by


G^i(x,v)�:= \frac{g^{ij}(x,v)}{4}\left(2\frac{\partial g_{jk}}{\partial x^\ell}(x,v) - \frac{\partial g_{k\ell}}{\partial x^j}(x,v)\right)v^k v^\ell.

The vector field H on TM/0 satisfies JH=V and [V,H]=H, where J and V are the canonical endomorphism and the canonical vector field on TM \0. Hence, by definition, H is a spray on M. The spray H defines a nonlinear connection on the fibre bundle TM \0 → M through the vertical projection

 v:T(TM\setminus 0)\to T(TM\setminus 0) \quad�; \quad v�:= \tfrac{1}{2}\big( I %2B \mathcal L_H J \big).

In analogy with the Riemannian case, there is a version

D_{\dot\gamma}D_{\dot\gamma}X(t) %2B R_{\dot\gamma}(\dot\gamma(t),X(t)) = 0

of the Jacobi equation for a general spray structure (M,H) in terms of the Ehresmann curvature and nonlinear covariant derivative.

Uniqueness and minimizing properties of geodesics

By Hopf–Rinow theorem there always exist length minimizing curves (at least in small enough neighborhoods) on (MF). Length minimizing curves can always be positively reparametrized to be geodesics, and any geodesic must satisfy the Euler-Lagrange equation for E[γ]. Assuming the strong convexity of F2 there exists a unique maximal geodesic γ with γ(0) = x and γ'(0) = v for any (xv) ∈ TM \ 0 by the uniqueness of integral curves.

If F2 is strongly convex, geodesics γ : [0, b] → M are length-minimizing among nearby curves until the first point γ(s) conjugate to γ(0) along γ, and for t > s there always exist shorter curves from γ(0) to γ(t) near γ, as in the Riemannian case.

Notes

  1. ^ G. Randers (1941), Phys. Rev. 59, 195.

References

External links