Method for applying adaptive control theory to quantum systems

Holland, Nathan David

Published Jan 1, 2015• 0 citations

Abstract

Control in quantum systems is proving to be one of the top challenges in the computing world as scalability of qubits in quantum computers increase. This thesis has focused on understanding quantum control methods and contribute further understanding to progress this very important step in advancing computers. By applying principles from adaptive and Lyapunov-based control to the effect on particle states within a closed quantum system as defined by the Schrödinger equation, it will be possible to see the precise boundaries of creating various control laws as states change over time. The benefits for these methods will be used in trajectory tracking and state targeting as applied to Schrödinger’s time dependent equation and simulated with pre-defined evolving Hamiltonian eigenstates to create reference and error control between target and system states. The analysis of this closed quantum system will be useful in applying similar methods in more tedious, externally influenced, open quantum systems.

Related papers

Lyapunov-Based Stabilization and Control of Closed Quantum Systems

Elham Jamalinia, Peyman Azodi, Alireza Khayatian +1 more

A Lyapunov-based method is presented for stabilizing and controlling of closed quantum systems. The proposed method is constructed upon a novel quantum Lyapunov function of the system state trajectory tracking error. A positive-definite operator in the Lyapunov function provides additional degrees of freedom for the designer. The stabilization process is analyzed regarding two distinct cases for this operator in terms of its vanishing or non-vanishing commutation with the Hamiltonian operator of the undriven quantum system. To cope with the global phase invariance of quantum states as a result of the quantum projective measurement postulate, equivalence classes of quantum states are defined and used in the proposed Lyapunov-based analysis and design. Results show significant improvement in both the set of stabilizable quantum systems and their invariant sets of state trajectories generated by designed control signals. The proposed method can potentially be applied for high-fidelity quantum control purposes in quantum computing frameworks.

arXiv:2102.00344Jan 31, 20210 citations

Dynamical trajectory tracking control in quantum systems

Zhu Yaping, Shuang Cong, Jianxiu Liu

Based on model reference adaptive control theory, we design a control law for a given closed quantum system model by means of the Lyapunov stability theorem by choosing a time-varying objective function as the reference model. Meanwhile, the problems of singularities and a very large value of the control fields during the tracking are solved by using an adaptive tracking algorithm. The simulation experiments realize the tracking task from the arbitrary initial states to the time-varying objective function. Under the effect of control law, the state of control system can track from the arbitrary initial state to the state of objective system. Simulation results demonstrate the ability of the control method proposed to adaptively adjust the control laws and real-time track dynamic target system.

Chinese Control Conference (CCC)Jul 25, 20120 citations

Lyapunov Control of Quantum Systems with Applications to Quantum Computing

K. P. Nagarjun, S. Sivaranjani, George Koshy

In the design of complex quantum systems like ion traps for quantum computing, it is usually desired to stabilize a particular system state or make the system state track a desired trajectory. Several control theoretical approaches based on feedback seem attractive to solve such problems. But the uncertain dynamics introduced by measurement on quantum systems makes the synthesis of feedback control laws very complicated. Although we have not explicitly modeled the change in system dynamics due to measurement (we have assumed weak measurements), this is a first step towards a more detailed analysis and closed-loop feedback design. Here, we present a Lyapunov-based control approach on the lines of that developed by Mirrahimi, Rouchon, Turnici (2005). The states are assumed to be obtained from weak measurements. The Lyapunov control technique has not been applied to realistic quantum systems so far. We have extended and applied the technique to two realistic physical systems - the quantum harmonic oscillator and the n-qubit system. We also propose to extend this concept to ion traps.

arXiv:1210.8293Oct 31, 20120 citations

Adaptive trajectory tracking of quantum systems

Jianxiu Liu, Shuang Cong, Zhu Yaping

A control strategy for tracking the trajectory of a time-variant objective function in closed quantum systems is proposed in this paper. The Lyapunov stability theorem is used to design the control law. Meanwhile, the issue of singularity is discussed and the problem of large value of control field occurred during the process of tracking are solved by using an adaptive regulation algorithm of objective function. Simulation results demonstrate the effectiveness and feasibility of the method proposed. We also explore the relationship of effect between control parameters and the error of the control system.

International Conference on Control, Automation and Systems (ICCAS)Dec 31, 20124 citations

Adaptive Quantum Control Theory

Jincheng Zhang

This paper introduces Adaptive Quantum Control Theory (AQCT), a novel framework designed to address the challenges inherent in controlling quantum systems. Traditional quantum control methods often rely on precise, pre-determined control signals, which can be highly sensitive to system imperfections and environmental noise. AQCT leverages the principles of reinforcement learning and quantum control theory to create a controller capable of dynamically adapting its control strategy based on the real-time state of the quantum system. The core mechanism involves designing a controller that learns and optimizes control policies through iterative interaction with the system, effectively mitigating the impact of uncertainties. This approach promises improved control performance, robustness, and scalability for a wide range of quantum systems. The theoretical framework presented here lays a foundation for future research and development in adaptive quantum control.

Aug 31, 20260 citations

Control and filtering of quantum systems

Shizhou Fu

This thesis is concerned with the control and filtering problems for quantum systems. Two kinds of control problems are considered for a class of quantum systems: optimal control and trajectory tracking control. Filtering of a class of linear quantum systems is also investigated. \n \nFor the optimal control of quantum systems, the system in question evolves on a manifold inR^3 and is modeled as a bilinear control form whose states are represented as coherence vectors. An associated matrix Lie group system with state space S O(3) is introduced in order to facilitate solving the given problem. The controllability as well as the reachable set of the system is analyzed in detail. The maximum principle is applied to the optimal control for systems evolving on the Lie group of special orthogonal matrices of dimension 3, with cost that is quadratic in the control input. Controlling the evolutions of coherence vector is equivalent to controlling the states of an associated matrix Lie group system. Thus, the optimal control problem for single spin-1/2 systems can be transformed to steer state of the matrix Lie group system from the identity matrix to a final matrix corresponding to a target state of single spin-1/2 systems in an optimal fashion. The obtained result is employed to perform a reversible logic quantum operation NOT on single spin-1/2 systems. Explicit expressions for the optimal control are given which are linked to the initial state of the system. The main contribution of the result is to extend the existing result to a larger class of systems, generalizing existing wave-function-based results to single spin-1/2 systems whose states are represented by density matrices such that it allows the inclusion of mixed states. \n \nFor the tracking control problem of closed single spin-1/2 quantum ensembles, based on the Liouville-von Neumann equation, mathematical model of dynamics for single spin-1/2 quantum ensembles are formulated as a bilinear form with states evolving on Bloch sphere. For a typical setting in NMR, one in general would not cancel the applied electromagnetic field along a specified direction, thus leading to the dynamics in a plane perpendicular to that direction on the Bloch sphere. One natural question arises that how can one design a feedback controller such that given an initial condition for quantum ensembles, the system evolutions can be steered to the pre-determined trajectory. An associated matrix Lie group differential equation describing transfers from the initial state being an identity matrix is naturally introduced. The tracking problem for the system of interest is transformed to a stabilization problem such that the control techniques are able to be used. A distance function as well as a distance vector defined on the Lie group SO(3) are introduced, which can be used to construct a Lyapunov function to deal with the tracking problem. In order to deal with the tracking problem, the Lyapunov-based feedback design method is employed to obtain the feedback control such that the dynamics of system can converge to the reference trajectory. \n \nFor filtering of quantum systems, the filter problem of a class of linear quantum stochastic systems with quantum noise is investigated. The quantum system model is first analyzed, then the necessary conditions of nondemolition are obtained by applying the observation principle. The relations between the input and output noise are necessary for the nondemolition measurement of the quantum output channels. Given prescribed quantum system, the filter of the system is derived using those necessary conditions.

Jan 1, 20150 citations