The N-Body Problem on Coadjoint Orbits
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematical Physics. The day in mathematical physics: rigorous models of physical systems.
Kai: We're looking at the paper: "The N-Body Problem on Coadjoint Orbits".
Mira: Here's the gist of the paper. The paper shows that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions because it relates certain singular symplectic spaces to (Marsden-Weinstein-Meyer) symplectic reduced spaces for the n-body problem.
Kai: Lax Pair Representation and Reduction Newton’s equations for n bodies moving in Euclidean space R d form a set of d×n second order differential equations which can be rewritten as a system of first order Hamiltonian equations on a phase space Pd,n of dimension 2nd.
Mira: This system is encoded by the evolving matrix K(t) ∈ g = sp(2n − 2) of (1) which encodes the “Galilean shape” of a moving phase point.
Kai: The quotient space Pd,n by the action of the Galilean group is identified with the Poisson variety symm+(2n − 2; d) ⊂ sp(2n − 2)∗ with corresponding quotient map G ◦ πtrans.
Paper summary: Kai: So to wrap up the paper "The N-Body Problem on Coadjoint Orbits," it establishes that the spatial n-body problem at non-zero angular momentum has a singular symplectic space split into two parts: spatial motions and planar motions.
Mira: They prove these strata are coadjoint orbits in the dual of Sp(2n minus two), and they show that this works even when you reduce by going to center-of-mass coordinates and then using the orthogonal group O for reduction <ref:2504.01272#pg1,reduce by going to center-of-mass coordinates>.
Lev: What this means for me on a real hardware side is that we need to be very careful about which reduction method we use, because the difference between O and SO leads to a singular space versus a smooth one.
Kai: And the main implication is that they provide a complete classification of these coadjoint orbits in low dimensions, which gives us tools to understand the dynamics better for n-body systems.
Mira: It's about finding the right mathematical language—the Lie-Poisson structure on the dual space—to describe these complex physical systems in a way that makes sense geometrically.
Conclusion: Kai: So we're wrapping up this look at "The N-Body Problem on Coadjoint Orbits." The authors, Dullin and Montgomery, they really map out how to use these coadjoint orbits to figure out the structure of n-body problems at non-zero angular momentum.
Mira: Yeah, it boils down to them showing that when you reduce the n-body problem—that big system of equations—by looking at its underlying mathematical symmetry, it doesn't just become one smooth space. It splits into two distinct pieces: one for spatial motion and one for planar motion.
Lev: From where I sit with error correction, that split is important because it tells us exactly where the dynamics live. If we're trying to build a simulator, knowing those two separate strata helps you figure out which mathematical space you need to define your state on.
Kai: Right, so the title isn't just fancy math talk; it’s about finding that split structure in the n-body problem when angular momentum is present. It connects classical mechanics directly to these algebraic objects called coadjoint orbits.
Mira: Exactly. The implication for us theorists is that this gives us a concrete way to classify all the possible reduced systems, which we can then test against physical configurations we actually see in space, like planetary orbits or even how gas clouds behave.
Lev: And for me, it’s about the structure of the solution space. They give you a way to handle those singularities that usually break your numerical code. It shows where the "bad" points are and how to navigate around them smoothly using this reduction method.
Kai: So basically, they've given us a map—a classification system—for these complex dynamical systems, whether we're talking about just three bodies or hundreds of them. It’s a powerful way to organize the chaos.
Mira: And it opens the door for finding new configurations that were previously hard to find because you had no idea which mathematical space to search in first. Next, we need to see how this algebraic structure actually translates into those specific physical orbits, and what happens when you move from three dimensions up to higher dimensions.
math-ph, math.DS, math.MP
Submitted: 2025-04-02
Updated: 2025-04-02
Comments: 28 pages, 2 figures
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 83/100
The gist: The paper shows that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata…
Key concepts
- Symplectic Reduction
- This is a mathematical technique used to simplify complex dynamical systems, like the n-body problem. It involves taking a large phase space and 'reducing' it by imposing constraints (like conserved quantities) from symmetry groups. The result is a smaller, simpler space that still captures the essential dynamics.
- Coadjoint Orbits
- These are specific paths or sets of points within a Lie algebra structure that are invariant under certain transformations. In this paper, these orbits represent the different possible reduced spaces for the n-body problem. The shape and size of these orbits determine the nature of the resulting symplectic space.
- Symplectic Strata
- These are distinct, separate regions within a singular symplectic space. The paper identifies two such strata: one corresponding to motions in 3D space and another corresponding to motions that are confined to a plane. This shows how different physical behaviors emerge from the same underlying mathematical structure.
Terminology
Summary
The paper shows that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions because it relates certain singular symplectic spaces to (Marsden-Weinstein-Meyer) symplectic reduced spaces for the n-body problem. p. 1 Our main result, theorem 2, relates certain
singular symplectic spaces consisting of the closure of certain coadjoint orbits in g
∗
to the
(Marsden-Weinstein-Meyer) symplectic reduced spaces for the n-body problem, reduced
at certain values of angular momentum. In particular, we show in theorem 3 that the
symplectic reduction of the spatial n-body problem with respect to the orthogonal group
O(3) at nonzero angular momentum consists of a singular space which is the union of two
g
∗
coadjoint orbits.Where the paper says it
Lax Pair Representation and Reduction
Newton’s equations for n bodies moving in Euclidean space R d form a set of d×n second order differential equations which can be rewritten as a system of first order Hamiltonian equations on a phase space Pd,n of dimension 2nd. This system is encoded by the evolving matrix K(t) ∈ g = sp(2n − 2) of (1) which encodes the “Galilean shape” of a moving phase point. p. 0 and that these in turn are equivalent to the Albouy-Chenciner version of the reduced
N-body equations described in their celebrated paper [2]. The evolving matrix K(t) ∈ g = sp(2n − 2) of (1) encodes the “Galilean shape” of a
moving phase point. Putting this together with some understanding of
Poisson maps we have:
Theorem 1. The quotient space of n-body phase space Pd,n by the action of the Galilean
group is the Poisson variety symm+(2n − 2; d) ⊂ sp(2n − 2)∗ = g
∗ with corresponding
quotient map G ◦ πtrans.Where the paper says it
Where the paper says it
Angular Momentum and Invariants
The evolving matrix K’s conserved eigenvalues correspond to the angular momentum L as a conserved quantity in the n-body problem. The spectral lemma shows that trL k = tr Kk, k = 0, 1, 2,. p. 5 More generally we have:
Lemma 1 (Spectral Lemma). trL
k = tr Kk
, k = 0, 1, 2, . Here we work out and record details of theorem 2
when d = 3 and L ̸= 0, that is, when the bodies move in standard three-space with non-zero
angular momentum. In this case the rank d of the motion space is given by d = 2p+q = 3.Where the paper says it
Where the paper says it
Structure of Coadjoint Orbits
The normal form (11) for elements in sp(2m)+ specifies the numerical invariants (p, q) and spectral invariants ω ij, j = 1,... p. The orbit is closed if and only if q = 0. p. 8 , p its spectral invariants. Then:
(i) q = 0 ⇐⇒ the orbit is closed
(ii) p = 0 ⇐⇒ L = 0
(iii) d ≤ 3 =⇒ p ≤ 1 and if p = 1 then ω
2
1 = ∥L∥
2
. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2).Where the paper says it
Where the paper says it
Resolution of Singularities
The singularity in the O(3) reduced space at non-zero angular momentum is resolved by reducing by the proper orthogonal group SO(3) instead of the full orthogonal group O(3). This leads to a smooth symplectic reduced space L−1(µ)/SO(2), which is everywhere smooth. p. 7 Let
Gµ ⊂ O(d) denote the isotropy group for µ, meaning the subgroup of all g ∈ O(d) such
that gµgt = µ. Then Gµ acts on L
−1
(µ) and, in ‘very nice’ situations L
−1
(µ) is a smooth
manifold, and this Gµ action sweeps out the kernel of the ambient symplectic form restricted
to L
−1
(µ) and the quotient space becomes a symplectic manifold: the symplectic reduced
space L
−1
(µ)/Gµ.Where the paper says it
General Results and Dimensions
The dimension of the phase space for n bodies moving in 3-space is 6n, and the image in sp(2n−2) whose rank is 3 or less has dimension 6n − 9. For large n, only the thinnest shell within the matrix space is relevant for the spatial n-body problem. p. 3 We then identify the symplectic reduced spaces for the
n-body problem in d-dimensions as closures of coadjoint orbits within this subvariety, see
theorem 2. Astronomers are most interested in the case d = 3 of the spatial three-body
problem but our constructions hold for any d. Since q = 0 corresponds to the normal
form being semi-simple we get q = 0 =⇒ the orbit is closed. We could use the same type
of argument to show q > 0 =⇒ not closed, but we prefer to verify this directly. , q, the choice q
′ = q
corresponding to the original orbit. The orbit’s closure realizes the symplectic reduction of
the m+ 1-body problem in R
d
reduced at L. , q, the choice q
′ = q
corresponding to the original orbit. The orbit’s closure realizes the symplectic reduction of
the m+ 1-body problem in R
d
reduced at L.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Appendix Details
Appendix A provides a singular-value-decomposition-like factorisation due to Xu [18] which shows that every matrix Z ∈ Pd,2n−2 can be written as Z = QDT−1. This normal form yields the quadratic Hamiltonian representation which is the most efficient way to read out the information contained in elements of sp(2n − 2). p. 5 As elements of o(d) we
have Xi ∧ Yi = XiY
t
i − YiXt
i
. A simple computation now yields equation (6) above for
angular momentum. The rank 2p + q is the rank of Z and equals the dimension of the affine space swept
out by the Newtonian solution for any initial condition (q, p) ∈ Pd,n which has has normal
form λ when this affine space is viewed from the center-of-mass frame . The coadjoint orbit through λ ∈ symm(2n − 2) is uniquely determined by the
numerical invariants and spectral invariants of µ.Where the paper says it
Where the paper says it
Low Dimensional Cases
For the 3-body problem, the coadjoint orbits live in the 10 dimensional dual of the Lie algebra sp(4). The dimension of the generic orbit is 8. p. 8 The rank 2p + q is the rank of Z and equals the dimension of the affine space swept
out by the Newtonian solution for any initial condition (q, p) ∈ Pd,n which has has normal
form λ when this affine space is viewed from the center-of-mass frame . The coadjoint orbit through λ ∈ symm(2n − 2) is uniquely determined by the
numerical invariants and spectral invariants of µ. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2). We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2). We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
References
[1] Alain Albouy. Integral manifolds of the n-body problem. p. 7 We still define the reduced
space to be L
−1
(µ)/Gµ but it is typically not a smooth manifold. Rather, the reduced
space is a singular symplectic manifold, meaning an analytic space which is stratified in
the Whitney sense and whose strata are smooth symplectic manifolds.Where the paper says it
[2] Alain Albouy and Alain Chenciner. p. 0 In section 6. 4 we show that this Lax pair representation of the
n-body equations is equivalent to the Albouy-Chenciner [2] representation in terms of
symmetric matrices.Where the paper says it
[3] Suntharan Arunasalam, Holger R Dullin, and Diana MH Nguyen. p. 0 THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. THE N-BODY PROBLEM ON COADJOINT ORBITS 5
the n-body problem to the regularised n-body problem is discussed, which leads to the
Lie-Poisson structure u(m, m), along with a Lax pair. Deriving the Lax Pair
Here we derive the Lax pair equation (1).Where the paper says it
Where the paper says it
[4] A. V. Bolsinov, A. V. Borisov, and I. S. Mamaev. Lie algebras in vortex dynamics and celestial mechanics IV Regul Chaos Dyn. p. 3 to achieve the arrow πtrans and work in centered
phase space. Dual pairs first appeared in representation theory and quantum mechanics.Where the paper says it
[5] M. Cerkaski. Application of Poisson maps on coadjoint orbits of Sp(6, R) group to many body dynamics J. p. 6 Symplectic Reduction and coadjoint orbits. The relationship between O(d) and Sp(2n − 2) and the dual nature of their momentum
maps lets us identify a symplectic reduced space for one group with the closure of a coadjoint
orbit for the other.Where the paper says it
[6] Alain Chenciner. p. 0 In section 6. 4 we show that this Lax pair representation of the
n-body equations is equivalent to the Albouy-Chenciner [2] representation in terms of
symmetric matrices. 4 we show that these Lax pair equations are equivalent to the
Albouy-Chenciner equations. Albouy and Chenciner [2] describe how Lagrange first reduced the spatial three-body
problem using invariant polynomials.Where the paper says it
Where the paper says it
[7] H. R. Dullin. The Lie-Poisson structure of the reduced n-body problem. p. 4 THE N-BODY PROBLEM ON COADJOINT ORBITS 5
the n-body problem to the regularised n-body problem is discussed, which leads to the
Lie-Poisson structure u(m, m), along with a Lax pair. Deriving the Lax Pair
Here we derive the Lax pair equation (1).Where the paper says it
[8] Roger Howe. p. 6 , Z2n−2 ∈ R
d
are the column vectors of Z. What Roger Howe has called “the first main theorem of
invariant theory” asserts that any polynomial function on Pn,d−1 which is invariant under
O(d) can be expressed as a polynomial in the Gij . The former equations automatically have the adjoint orbits as invariant
submanifolds while the latter set of equations have the corresponding image coadjoint orbits
of g
∗ as invariant symplectic submanifolds. Symplectic Reduction and New Results.Where the paper says it
Where the paper says it
[9] Roger Howe. p. 6 , Z2n−2 ∈ R
d
are the column vectors of Z. What Roger Howe has called “the first main theorem of
invariant theory” asserts that any polynomial function on Pn,d−1 which is invariant under
O(d) can be expressed as a polynomial in the Gij . , Z2n−2 ∈ R
d
are the column vectors of Z. What Roger Howe has called “the first main theorem of
invariant theory” asserts that any polynomial function on Pn,d−1 which is invariant under
O(d) can be expressed as a polynomial in the Gij . 1
arXiv:2504. 01272v1 [math-ph] 2 Apr 2025Where the paper says it
Where the paper says it
Where the paper says it
[10] Wu-yi Hsiang. Cohomology theory of topological transformation groups, volume Band 85 of Ergebnisse der Mathematik und ihrer Grenzgebiete [Results in Mathematics and Related Areas]. Springer-Verlag, New York-Heidelberg, 1975. p. 1 The former equations automatically have the adjoint orbits as invariant
submanifolds while the latter set of equations have the corresponding image coadjoint orbits
of g
∗ as invariant symplectic submanifolds. Symplectic Reduction and New Results.Where the paper says it
[11] Eugene Lerman, Richard Montgomery, and Reyer Sjamaar. p. 0 THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. 1
arXiv:2504. 01272v1 [math-ph] 2 Apr 2025 A simple computation now yields equation (6) above for
angular momentum. Note that tr(L
2
) = tr(ZJZtZJZt
) = tr(JZtZJZtZ) = tr(K2
) where
K = JG as above.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[12] R Moeckel. Lecture notes on central configurations (his website), 2016.
[13] Filomena Pacella. Central configurations of the N-body problem via equivariant Morse theory. p. 0 THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions.Where the paper says it
[14] S. T. Sadetov. On the regular reduction of the n-dimensional problem of N +1 bodies to Euler-Poincar´e equations on the Lie algebra sp(2N). p. 0 We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2).Where the paper says it
[15] Paul Skerritt and Cornelia Vizman. Dual pairs for matrix lie groups Journal of Geometric Mechanics, 11:255–275, 2019. p. 3 Subsequently the papers [4], [14], [5], [7] use dual
pairs in connection with the n-body problems. General facts about dual pairs for matrix
groups as they arise in geometric mechanics can be found in [15].Where the paper says it
[16] Hermann Weyl. The classical groups. p. 3 See Howe [8]
and [9]. That dual pairs arise in connection to the classical n-body problem was pointed out
in [11]. In this
incarnation the Lie bracket is the Poisson bracket and the coadjoint action is pullback:
λ 7→ λ ◦ g
−1 where λ ∈ g ∼= g
∗ and g ∈ Sp(2n − 2). Words are in order regarding going back and forth between our three representations
of sp(2n − 2).Where the paper says it
Where the paper says it
[17] Aurel Wintner. p. 4 We also show in section 6 that
(9) P = JS with J =
0 I
−I 0
and S =
M˜ −1 0
0 A˜
,
where the entries of the symmetric matrix S contain first derivatives of the Hamiltonian. The (n − 1) × (n − 1) symmetric matrix A˜ is a version of what Albouy-Chenciner call the
“Wintner-Conley matrix” and depends only on the small Gram matrix XtX. to achieve the arrow πtrans and work in centered
phase space. Dual pairs first appeared in representation theory and quantum mechanics.Where the paper says it
Where the paper says it
[18] Hongguo Xu. An svd-like matrix decomposition and its applications. p. 2 Angular momentum and spectral invariance. The Lax pair evolution (1), like
any Lax pair evolution, preserves the spectrum of the evolving matrix K, since the equation
says that K evolves by conjugation by some time-dependent symplectic matrix. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2).Where the paper says it
Where the paper says it
Appendix C Details
C.1 The 2-body problem occurs on the dual of the three-dimensional Lie algebra sp(2) = sl(2, R). p. 0 We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2). This dual pair
structure leads almost immediately to theorem 1 which identifies the Galilean quotient of
phase space as a Poisson subvariety of sp(2n − 2)∗
, specifically as the subvariety of all
positive semi-definite matrices G. We then identify the symplectic reduced spaces for the
n-body problem in d-dimensions as closures of coadjoint orbits within this subvariety, see
theorem 2. In order to reverse and go from K back to λ set λ(v) = 1
2
ω(v, Kv) where
ω =
Pdxi ∧ dyi
is the symplectic form. For the rest of this section we use the quadratic Hamiltonian realization of points of
sp(2n − 2). The orbit’s closure realizes the symplectic reduction of
the m+ 1-body problem in R
d
reduced at L. The (p, q′
) type orbits with q
′ < q correspond to
the projections to the reduced space of phase points Z ∈ Pd,m with rank(Z) = 2p + q
′ < d. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Each stratum
is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group
Sp(2n−2). THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. THE N-BODY PROBLEM ON COADJOINT ORBITS 3
The orthogonal group O(d) is the subgroup of the Galilean group which fixes the center
of mass 0 ∈ R
d and acts on centered phase space Pd,n−1, with g ∈ O(d) acting by Z 7→ gZ. The Gram matrix
(3) G = Z
tZ ∈ symm(2n − 2)
is invariant under this action, and is a complete invariant.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Conclusion
The paper establishes a framework for identifying symplectic reduced spaces for the n-body problem as closures of coadjoint orbits within a specific subvariety. p. 3 This dual pair
structure leads almost immediately to theorem 1 which identifies the Galilean quotient of
phase space as a Poisson subvariety of sp(2n − 2)∗
, specifically as the subvariety of all
positive semi-definite matrices G. We then identify the symplectic reduced spaces for the
n-body problem in d-dimensions as closures of coadjoint orbits within this subvariety, see
theorem 2. The reader will also find the
claim that this image is the closure of a single coadjoint orbit lying in sp(2n − 2)+, a fact
we will be later proving by hand. We recall the construction of an O(d) symplectic reduced space, also known as a
Marsden-Weinstein-Meyer reduced space. The reader will also find the
claim that this image is the closure of a single coadjoint orbit lying in sp(2n − 2)+, a fact
we will be later proving by hand. We recall the construction of an O(d) symplectic reduced space, also known as a
Marsden-Weinstein-Meyer reduced space. THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions. Let
Gµ ⊂ O(d) denote the isotropy group for µ, meaning the subgroup of all g ∈ O(d) such
that gµgt = µ. Then Gµ acts on L
−1
(µ) and, in ‘very nice’ situations L
−1
(µ) is a smooth
manifold, and this Gµ action sweeps out the kernel of the ambient symplectic form restricted
to L
−1
(µ) and the quotient space becomes a symplectic manifold: the symplectic reduced
space L
−1
(µ)/Gµ. THE N-BODY PROBLEM ON COADJOINT ORBITS 3
The orthogonal group O(d) is the subgroup of the Galilean group which fixes the center
of mass 0 ∈ R
d and acts on centered phase space Pd,n−1, with g ∈ O(d) acting by Z 7→ gZ. The Gram matrix
(3) G = Z
tZ ∈ symm(2n − 2)
is invariant under this action, and is a complete invariant. Date: starting date: May 2021; finished: January 2025? 1
arXiv:2504.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
--- Page 1 ---
THE N-BODY PROBLEM ON COADJOINT ORBITS HOLGER DULLIN & RICHARD MONTGOMERY Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions. p. 0 THE N-BODY PROBLEM ON COADJOINT ORBITS
HOLGER DULLIN & RICHARD MONTGOMERY
Abstract. We show (Theorem 3) that the symplectic reduction of the spatial n-body
problem at non-zero angular momentum is a singular symplectic space consisting of two
symplectic strata, one for spatial motions and the other for planar motions.Where the paper says it
Pruning redundant reasoning tokens
The authors show (Theorem 3) that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions.
How the training signal is built
The authors show (Theorem 3) that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions.
Where the gains come from
The authors show (Theorem 3) that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions.
Summary
The authors show (Theorem 3) that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions. The paper establishes a framework for identifying symplectic reduced spaces for the n-body problem as closures of coadjoint orbits within a specific subvariety. This construction is generalizable to any dimension d.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI system can perform symplectic reduction on n-body problems by identifying singular symplectic spaces as unions of coadjoint orbits in sp(2n-2). This allows for precise classification of reduced phase spaces based on angular momentum and rank, as shown in Theorem 3.
-
Improved AI system can utilize the dual pair perspective between the orthogonal group O(d) and the symplectic group Sp(2n-2). This enables the identification of momentum maps L (angular momentum) and G (Gram map), which are
complete invariants
for their respective groups, simplifying the analysis of reduced spaces. -
Improved AI system can construct a normal form for coadjoint orbits using Xu's singular value decomposition-like factorisation. This normal form allows for the unique determination of coadjoint orbits by numerical and spectral invariants, which directly maps to the structure of the reduced symplectic manifolds.
-
Improved AI system can differentiate between spatial and planar reduced spaces based on the rank of their points in Pd,n-1. It can predict that
No spatial (i.e. rank 3) solution can, in finite time, evolve into a planar (rank 2) phase point.
-
Improved AI system can resolve singularities in the O(d) reduced space by applying the SO(d) reduction instead of the full O(d). This leads to an everywhere smooth symplectic manifold for angular momentum L not equal to zero, as described in Theorem 4.
Abstract
We show (Theorem 3) that the symplectic reduction of the spatial n-body problem at non-zero angular momentum is a singular symplectic space consisting of two symplectic strata, one for spatial motions and the other for planar motions. Each stratum is realized as coadjoint orbit in the dual of the Lie algebra of the linear symplectic group Sp(2n-2). The planar stratum arises as the frontier upon taking the closure of the spatial stratum. We reduce by going to center-of-mass coordinates to reduce by translations and boosts and then performing symplectic reduction with respect to the orthogonal group O(3). The theorem is a special case of a general theorem (Theorem 2) which holds for the n-body problem in any dimension d. This theorem follows largely from a ``Poisson reduction'' theorem, Theorem 1. We achieve our reduction theorems by combining the Howe dual pair perspective of reduction espoused by Lerman-Montgomery-Sjamaar with a normal form arising from a symplectic singular value decomposition due to Xu. We begin the paper by showing how Poisson reduction by the Galilean group rewrites Newton's equations for the n-body problem as a Lax pair. In section 6.4 we show that this Lax pair representation of the n-body equations is equivalent to the Albouy-Chenciner representation in terms of symmetric matrices.
Related papers
- Existence of a fluid phase for the two-dimensional one-component plasma
- The Huang-Yang conjecture for the low-density Fermi gas
- Anderson localisation in infinite and finite resonator crystals with long-range interactions
- Free energy and spectral edge of the SYK model
- Anderson localization for 1-d quasi-periodic Schr"odinger operators with degenerate weights
- Irregular SLE(4) martingales and isomonodromic deformations