File pdf 360KB

Introduction to Quantum Control with a Focus on Measurement Feedback Quantum Control

Thermostats, washing machines, robotic systems, and many other consumer and industrial products are regulated with control theory to automate their tasks. These devices can follow prescribed routines as well as compensate for environmental fluctuations through sensing of variables of the system. However, traditional control theory, which often involves taking measurements of relevant quantities in order to correct deviations from the desired behavior, does not transfer directly to quantum systems. Quantum systems may be as small as a single hydrogen atom and are comprised of subatomic particles subject to non-negligible quantum effects due to their small size. These systems are inherently affected by the action of taking a measurement through Heisenberg’s Uncertainty Principle. Noise is introduced into the system as the increase in certainty of one variable caused by a measurement necessarily decreases the certainty of a complementary variable’s state. While open-loop quantum control, which avoids taking measurements, is similar to traditional open-loop control, closed-loop quantum control is very different than classical closed-loop control because of the noise created by taking measurements. Approaches that are currently being refined for closed-loop quantum control include closed-loop learning control, direct feedback control, indirect feedback control, quantum filtering, and coherent quantum feedback. These methods provide ways to control some quantum systems while more methods and alterations are being created to work on other types of quantum systems. Many of these quantum control techniques are inspired by similar approaches in traditional control but are altered to apply to quantum systems. Quantum control aims to provide ways of controlling all types of quantum systems through certain combinations of new and old control strategies. This paper provides an introductory survey of the current state of quantum control and the progress made to this point.