arXiv:2609.18195
Redesigning the linear--quadratic--Gaussian cost function for feedback cooling of a quantum harmonic oscillator
Shuma Sugiura, Koki Shiraishi, Masahito Ueda
Abstract
Linear--quadratic--Gaussian (LQG) control is optimal only with respect to a prescribed cost function, the choice of which dictates the physical objective of the control. We consider feedback cooling of a continuously monitored quantum harmonic oscillator by shifting the minimum of its trapping potential. In this setting, the physically relevant cooling objective can be defined as minimizing the oscillator's energy relative to the feedback-shifted potential. In contrast, conventional LQG control evaluates the energy from a fixed origin and thus fails to directly optimize this quantity. To address this problem, we introduce a redesigned cost function that explicitly accounts for the feedback-induced shift of the potential. We then derive the corresponding optimal feedback law and obtain an analytic expression for the minimum achievable steady-state phonon occupation number. The redesigned LQG control achieves a lower occupation number than low-pass-filter (LPF) feedback formulated for the same cooling objective. While this improvement is minor at detection efficiencies currently attainable in experiments---indicating that LPF feedback already delivers near-optimal cooling performance---the advantage becomes pronounced as the detection efficiency approaches unity. In this regime, the redesigned LQG control provides an increasing advantage for reaching the motional ground state at a finite measurement strength. We clarify that the conventional and redesigned cost functions represent distinct control objectives rather than different implementations of the same optimization problem.
AI Summary
The paper considers measurement-based cooling of a quantum harmonic oscillator by shifting its trap minimum. It asks how to choose an LQG cost that minimizes energy relative to the feedback-shifted potential, rather than energy relative to a fixed origin with a separate control penalty. The proposed cost integrates $\langle \hat p^2/(2m)+m\omega^2(\hat x-X)^2/2\rangle$. Solving a classical cross-weighted linear–quadratic regulator problem and applying certainty equivalence gives the stationary feedback law $X=\langle\hat x\rangle-\langle\hat p\rangle/(m\omega)$. The conditional means are obtained with a quantum Kalman filter. For Gaussian states under continuous position measurement, the covariance equations close and are independent of the displacement control; their stationary solution, together with the conditional-momentum dynamics, yields an analytic minimum occupation at fixed measurement strength and detection efficiency. Writing $g=4\hbar k/(m\omega^2)$, the weak-measurement result is $n=n_0+\sqrt{\eta}g^2/16+O(g^3)$, where $n_0=(\eta^{-1/2}-1)/2$. Optimized low-pass-filter feedback has twice the leading excess occupation, $n_{\rm LPF}=n_0+\sqrt{\eta}g^2/8+O(g^4)$. Both approach the same efficiency-dependent bound as $g\to0$. The difference is small at the representative efficiency $\eta=0.3$ and becomes more appreciable near unit efficiency. The analysis assumes Gaussian initial states and sufficiently long control horizons. Practical limits on trap displacement or localization require additional penalties or constraints. The two cost functions serve distinct physical objectives; which better supports subsequent nonclassical-state preparation remains unresolved.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 9 pages, 2 figures