Two-Line-Center Problem

• The two-line-center problem is a geometric optimization and dynamical system challenge that minimizes the maximum distance from points to two fixed lines.
• Efficient algorithms using parametric search, convex hulls, and ε-certificates achieve near-linear approximations with optimal complexity bounds.
• Applications span quantum scattering, symbolic dynamics in two-center potentials, and cold-atom experiments, driving further research in high-dimensional extensions.

The two-line-center problem refers to a class of geometric optimization problems and dynamical systems centered on the interaction of points, particles, or field sources with two fixed lines or centers. In computational geometry, it denotes finding two lines that minimize the maximum distance from any point in a planar set to its closest line; in classical and quantum mechanics, it underlies tractable models for particle motion, scattering, or symbolic coding of orbits in two-center potential fields. Recent work provides both efficient algorithms for geometric covering variants and deep structural results for symbolic dynamics of integrable two-center systems. This article synthesizes key theoretical frameworks, algorithmic approaches, and physical models as they pertain to the two-line-center and two-center problems.

1. Formal Problem Definitions and Geometric Two-Line-Center Variants

The classical two-line-center problem considers a finite point set $P \subset \mathbb{R}^2$ and seeks two lines $L_1, L_2$ that minimize

$$r^* = \min_{L_1, L_2} \max_{p \in P} \min \{\operatorname{dist}(p, L_1), \operatorname{dist}(p, L_2)\}$$

This may be equivalently formulated as covering $P$ with the union of two strips (slabs) defined by $L_1, L_2$, and minimizing the largest strip width. Extensions include:

• Weighted variant: Each point $p_i$ has weight $w_i$, and distance is scaled accordingly (Bhattacharya et al., 2015).
• $k$-center variant: Place up to $k$ centers constrained to two lines, minimize the maximal (weighted) distance (Bhattacharya et al., 2015).
• Orientation-constrained variants: Lines required to be parallel, perpendicular, or have specified orientations/angles (Ahn et al., 2024, Chung et al., 7 Jan 2026).
• Approximation and parametric algorithms: Efficient computation for exact or $(1+\varepsilon)$-approximate solutions (Chung et al., 7 Jan 2026).

2. Algorithmic Results and Complexity Bounds

Advancements in exact and approximate algorithms have provided nearly optimal solutions for the two-line-center and its constrained forms:

Variant	Best Exact Time	Best Approx. Time	Key Techniques
Unconstrained	$O(n^2\log^2 n)$	$O((n/\varepsilon)\log(1/\varepsilon))$ (Chung et al., 7 Jan 2026)	Parametric search, certificates, anchor pairs
Two fixed orientations	$O(n\log n)$ (Ahn et al., 2024)	$O(n+1/\varepsilon)$ (Chung et al., 7 Jan 2026)	Bipartition sweep, width queries, pre-sorting
One fixed orientation	$O(n\log^3 n)$ (Ahn et al., 2024); $O(n\log n)$ (Chung et al., 7 Jan 2026)	$O(n+1/\varepsilon^2\log(1/\varepsilon))$	Dynamic hulls, parametric search, certificates
Fixed angle of intersection	$O(n^2\alpha(n)\log n)$ (Ahn et al., 2024)	$O(n+1/\varepsilon^3\log(1/\varepsilon))$	Rotational sweeps, envelope algorithms
Parallel/perpendicular lines	$O(n\log^2 n)$ (Bhattacharya et al., 2015)	$O(n/\varepsilon)$ (Chung et al., 7 Jan 2026)	Greedy, interval trees, center-square split

Central to many approaches is the notion of $\varepsilon$-certificate: a small representative subset guaranteeing that expanding width by $(1+\varepsilon)$ recovers global coverage, enabling reductions from $O(n^2\log^2n)$ exact methods to linear or near-linear $(1+\varepsilon)$-approximation (Chung et al., 7 Jan 2026).

3. Data Structures and Parametric Optimization

Algorithms for the two-line-center problem frequently exploit advanced data structures:

• Convex hull edge-trees for dynamic width queries over constrained orientations (Ahn et al., 2024).
• LIFO dynamic-width structures enabling efficient feasibility checks (Ahn et al., 2024).
• Interval trees and shortest-interval trees for disk covering and $k$-center variants on two lines (Bhattacharya et al., 2015).
• Window-sliding convex-hull algorithms for streaming linear-time approximations (Chung et al., 7 Jan 2026).

Parametric search techniques (e.g., Cole's speed-up of the AKS network) enable optimization over radius or width values in $O(n\log^2 n)$ or better (Bhattacharya et al., 2015), with specific techniques for narrowing to minimal feasible intervals and dynamic candidate generation.

4. Symbolic Dynamics and Hamiltonian Two-Center Systems

The Euler two-fixed-center problem models a mass in the plane attracted to two fixed sources, with Hamiltonian

$$H(x, y, p_x, p_y) = \frac{1}{2}(p_x^2 + p_y^2) - \frac{m_1}{\sqrt{(x+d)^2+y^2}} - \frac{m_2}{\sqrt{(x-d)^2+y^2}}$$

Integrability yields additional first integral $G$; regularization is achieved via elliptic–cylindrical coordinates and time rescaling (Dullin et al., 2015).

Symbolic dynamics: Orbits are coded by syzygy sequences, marking crossings of intervals between centers. These sequences correspond to Sturmian words determined by the slope $W=T_\nu/T_\lambda$ of linear flow on invariant Liouville tori. All regular orbits are bijective to a Sturmian word of rational slope, and finite Sturmian subwords encode "collision–collision" orbits. Combinatorial properties—including period lengths, continued-fraction interpretation, and grid-cutting construction—are characterized completely (Dullin et al., 2015).

5. Quantum Two-Center Scattering and Resonance Phenomena

In quantum waveguide settings, the two-center problem models scattering of a confined particle (mass $\mu$) by two fixed impurities at $z = \pm d/2$ under transverse harmonic confinement. The Hamiltonian is

$$H_{3D} = -\frac{\hbar^2}{2\mu}\nabla^2 + \frac{1}{2}\mu\omega_\perp^2\rho^2 + V_\text{pseudo}(r)$$

with $V_\text{pseudo}$ a regularized zero-range pseudopotential acting at the two centers (Shadmehri et al., 2018).

Scattering solutions are expanded in transverse oscillator eigenmodes; open and closed channels are treated with envelope-matching at the impurity locations using a mixed-derivative regularization. Scattering amplitudes $f_e$, $f_o$ (even, odd) are given by explicit formulas involving sums $\Lambda_0$, $\Lambda_d$, $F_d$. Unitarity and resonance conditions yield transcendental equations for the CIR position in $a_\perp/a_{3D}$. Notably, the two-center setup splits the standard Olshanii CIR into even and odd branches, with positions sensitive to separation $d$.

Effective 1D Hamiltonian and coupling constants $g_{1D}^\pm$ can be extracted, identifying reflection and transmission behaviors. The approach generalizes to $N$ impurities by symmetrizing the regularization operator. Physical interpretation connects directly to cold-atom experiments, including systems with trapped ions or Rydberg atoms and Feshbach tuning (Shadmehri et al., 2018).

6. Extensions, Limitations, and Open Directions

• Weighted two-perpendicular-line $k$-center remains open for $o(n\log n)$ decision algorithms (Bhattacharya et al., 2015).
• Complexity lower bounds are established: one-fixed-orientation exact two-line-center cannot be solved in $o(n\log n)$ algebraic-decision-tree time (Chung et al., 7 Jan 2026).
• High-dimensional generalizations and connections to multi-center symbolic dynamics or quantum scattering with more impurities are natural directions given current regularization techniques (Shadmehri et al., 2018).
• Certificate-based reductions and anchor-pair-based orientation sampling suggest further improvements for near-linear multicenter or multi-slab covering.

This synthesis highlights the deep interplay between geometric algorithms, physical models, and symbolic-dynamic codings in two-line-center and two-center problems, with recent work providing both optimal complexity bounds and novel structural insights.