Small-time approximate controllability of the logarithmic Schrödinger equation

Basic Information

• Paper ID: 2510.14461
• Title: Small-time approximate controllability of the logarithmic Schrödinger equation
• Authors: Karine Beauchard, Rémi Carles, Eugenio Pozzoli
• Classification: math.AP (Mathematical Analysis), math-ph (Mathematical Physics), math.MP (Mathematical Physics), math.OC (Optimal Control), quant-ph (Quantum Physics)
• Publication Date: October 16, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2510.14461

Abstract

This paper considers the Schrödinger equation with logarithmic nonlinearity and bilinear control on $\mathbb{T}^d$ or $\mathbb{R}^d$, proving small-time global $L^2$-approximate controllability. The proof extends the method introduced by the first and third authors in the linear case to the nonlinear framework by combining small-time controllability of phases and gradient flows. Due to the presence of nonlinearity, the required estimates are more difficult to establish than in the linear case. The proof is inspired by WKB analysis. This is the first result on global (small-time) approximate controllability for nonlinear Schrödinger equations with bilinear control.

Research Background and Motivation

Problem Background

1. Quantum Control Theory: The mathematical control theory of bilinearly controlled Schrödinger PDEs has undergone tremendous development over the past two decades, with theoretical questions originating from applications of quantum control in physics and chemistry (such as absorption spectroscopy) and computer science (such as quantum computing).
2. Topological Obstacles to Exact Controllability: In the linear case, the system is not exactly controllable due to the reachable set having empty interior in $S \cap H$. This topological obstacle persists in Schrödinger equations with polynomial nonlinearities.
3. Physical Significance of Small-time Controllability: Small-time controllability has particularly relevant physical significance because quantum systems, once constructed, have very short lifetimes before decay (such as through spontaneous photon emission), and the ability to control them in minimal time is an open challenge in physics.

Research Motivation

1. Overcoming Existing Limitations: Existing linear Schrödinger equations can achieve large-time approximate controllability, but small-time controllability faces obstacles.
2. Advantages of Nonlinearity: The logarithmic Schrödinger equation has advantages over polynomial nonlinearities: it is directly well-posed in $L^2$ for arbitrary spatial dimensions, and solutions are globally Lipschitz continuous with respect to initial conditions.
3. Filling Theoretical Gaps: This is the first result on small-time global approximate controllability for nonlinear Schrödinger equations with bilinear control.

Core Contributions

1. First proof of small-time global approximate controllability for nonlinear Schrödinger equations: For systems with logarithmic nonlinearity and bilinear control, this is a breakthrough result in the field.
2. Methodological Innovation: Successfully extends the control method from the linear case to the nonlinear framework, combining small-time controllability of phases and gradient flows.
3. Technical Breakthrough: Uses estimates inspired by WKB analysis to handle technical difficulties arising from nonlinearity.
4. Complete Theoretical Framework: Establishes a comprehensive theoretical system including well-posedness theory, representation formulas, phase control, and flow control.
5. Negative Results: Proves that certain systems lack controllability, providing boundary conditions for the theory.

Detailed Methodology

Problem Formulation

Consider the logarithmic Schrödinger equation of the form: $\left\{ \begin{aligned} &\left(i\partial_t + \frac{1}{2}\Delta - V\right)\psi(t,x) = \lambda\psi \log|\psi|^2(t,x) + \sum_{j=1}^m u_j(t)W_j(x)\psi(t,x)\\ &\psi(0,\cdot) = \psi_0 \end{aligned} \right.$

where $M$ is $\mathbb{R}^d$ or a smooth connected boundaryless Riemannian manifold, $V, W_1, \ldots, W_m$ are real-valued potential functions, and $u_1, \ldots, u_m$ are piecewise constant control functions.

Main Systems

The paper studies two concrete systems:

System 1 ($M = \mathbb{T}^d$): $\left(i\partial_t + \frac{1}{2}\Delta - V\right)\psi = \lambda\psi \log|\psi|^2 + \sum_{j=1}^d (u_{2j-1}(t)\sin + u_{2j}(t)\cos)\langle b_j, x\rangle)\psi$

System 2 ($M = \mathbb{R}^d$): $\left(i\partial_t + \frac{1}{2}\Delta - V\right)\psi = \lambda\psi \log|\psi|^2 + \left(\sum_{j=1}^d u_j(t)x_j + u_{d+1}(t)e^{-|x|^2/2}\right)\psi$

Core Strategy

The proof strategy is based on the following criteria:

1. Small-time approximate reachability maps ($L^2$-STAR maps): Describe states that can be reached by system trajectories in arbitrarily small time
2. Small-time controllability of phases (STC of phases): For any initial condition $\psi_0 \in S$ and phase $\varphi \in L^2(M,\mathbb{R})$, one can approximately reach the state $e^{i\varphi}\psi_0$ in arbitrarily small time
3. Small-time controllability of flows (STC of flows): Controllability of flows of gradient vector fields

Technical Innovations

1. Application of the Eikonal Equation

For at most linear $\varphi \in C^{\infty}(M,\mathbb{R})$, consider the eikonal equation: $\left\{ \begin{aligned} &\partial_s\phi(s,x) + \frac{1}{2}|\nabla\phi(s,x)|^2 = 0\\ &\phi(0,\cdot) = \varphi \end{aligned} \right.$

2. Representation Formula

Through time rescaling $\xi(s,x) := \psi(\tau s, x)$, obtain the representation formula: $\psi(\tau s; 0, \psi_0 e^{i\varphi/\tau}) = a(s,\cdot)e^{i\phi(s,\cdot)/\tau}$ where $a$ satisfies a specific evolution equation.

3. Special Properties of Logarithmic Nonlinearity

Key Lemma (Lemma 2.1): For $z_1, z_2 \in \mathbb{C}$, $\left|\text{Im}\left((z_2-z_1)(z_2\log|z_2|^2 - z_1\log|z_1|^2)\right)\right| \leq 2|z_1-z_2|^2$

This estimate is the central tool for handling logarithmic nonlinearity.

Experimental Setup

Theoretical Verification Framework

The paper is primarily theoretical work, with results verified through:

1. Well-posedness Proofs: Proving well-posedness of the Cauchy problem in $L^2(M)$
2. Convergence Analysis: Proving convergence of various approximations
3. Trotter-Kato Formula: Establishing convergence of splitting methods in the nonlinear case

Main Theorems

Theorem 1.4: Let $d \in \mathbb{N}^*$, $\lambda \in \mathbb{R}$. If $V \in L^{\infty}(\mathbb{T}^d,\mathbb{R})$, then system (1.3) is small-time $L^2$-approximately controllable.

Theorem 1.5: Let $d \in \mathbb{N}^*$, $\lambda \in \mathbb{R}$. If $V$ satisfies at most quadratic growth conditions, then system (1.5) is small-time $L^2$-approximately controllable.

Experimental Results

Positive Results

1. Small-time Controllability: Successfully proves small-time $L^2$-approximate controllability for the two main systems
2. Phase Control: Proves small-time controllability of phases (Theorem 3.1)
3. Flow Control: Establishes small-time controllability of gradient vector field flows

Negative Results

Theorem 1.6 provides important negative results:

1. For quadratic potentials $V(x) = \alpha|x|^2 + \beta \cdot x + \gamma$, system (1.8) is not (large-time) $L^2$-approximately controllable
2. For $V$ satisfying at most quadratic growth conditions, system (1.7) is not small-time $L^2$-approximately controllable

Key Estimates

The paper establishes several important estimates:

• Lipschitz continuity: $\|\psi(t;u,\psi_0) - \psi(t;u,\tilde{\psi}_0)\|_{L^2} \leq e^{2|\lambda|t}\|\psi_0 - \tilde{\psi}_0\|_{L^2}$
• Convergence rate estimates: Specific convergence rates are established in various approximations

Related Work

Exact Controllability

• Pioneering work by Ball, Marsden, and Slemrod showing topological obstacles in the linear case
• Exact controllability results on one-dimensional intervals (Beauchard, Coron, et al.)
• Topological obstacles persist in the polynomial nonlinearity case

Approximate Controllability

• Various methods for discrete spectrum cases: Galerkin approximation, periodic control laws, adiabatic approximation, Lyapunov stabilization
• Large-time approximate controllability results for other PDEs

Small-time Controllability

• Previously only small-time controllability results for linear equations existed
• Breakthrough work by Beauchard-Pozzoli in the linear case
• Applications of Agrachev-Sarychev saturation arguments in various PDEs

Conclusions and Discussion

Main Conclusions

1. First proof of small-time global $L^2$-approximate controllability for Schrödinger equations with logarithmic nonlinearity and bilinear control
2. Successfully extends the control method from the linear case to the nonlinear framework
3. Establishes a complete theoretical framework including well-posedness, representation formulas, and control strategies

Limitations

1. Method Limitations: The proof method is specific to logarithmic nonlinearity and cannot be directly generalized to other types of nonlinearity
2. Technical Requirements: Requires potential functions to satisfy specific regularity conditions (boundedness or at most quadratic growth)
3. Control Form: Limited to specific forms of bilinear control

Future Directions

1. Extension to Other Nonlinearities: Investigate whether the method can be generalized to other types of nonlinearity
2. More General Potential Functions: Consider more general classes of potential functions
3. Exact Controllability: Study exact controllability problems for logarithmic Schrödinger equations
4. Numerical Implementation: Develop numerical control algorithms based on theoretical results

In-depth Evaluation

Strengths

1. Theoretical Breakthrough: This is the first result in the field with pioneering significance
2. Methodological Innovation: Successfully applies WKB analysis techniques to control theory with novel technical approach
3. Completeness: Provides a complete theoretical framework from well-posedness to control strategies
4. Balance: Provides both positive and negative results, offering a complete picture of the theory

Weaknesses

1. Limited Application Scope: Results are specific to logarithmic nonlinearity with limited generalizability
2. Non-constructive: Proofs are existence-based without providing explicit control construction algorithms
3. Lack of Numerical Verification: Absence of numerical experiments to verify practical utility of theoretical results

Impact

1. Academic Value: Opens new directions in nonlinear quantum control theory
2. Technical Contribution: Established techniques may apply to other nonlinear PDE control problems
3. Physical Significance: Provides theoretical foundation for rapid control of quantum systems

Applicable Scenarios

1. Quantum Optics: Optical systems with logarithmic saturation nonlinearity
2. Bose-Einstein Condensates: Certain types of BEC systems
3. Quantum Mechanics: Modeling quantum systems with logarithmic nonlinearity

References

The paper cites 71 related references covering important works in control theory, partial differential equations, quantum mechanics, and numerical analysis, particularly:

• Pioneering work by Ball, Marsden, and Slemrod on bilinear system controllability
• Series of works by Beauchard et al. on Schrödinger equation control
• Research by Carles et al. on well-posedness of logarithmic Schrödinger equations
• Contributions by Agrachev and Sarychev on saturation arguments

This paper represents significant progress in nonlinear quantum control theory. By cleverly combining WKB analysis and control theory techniques, it resolves a challenging mathematical physics problem and lays the foundation for further development in the field.