Abstract

Quantum optimal control has evolved into one of the cornerstones for enabling quantum technologies.The computation of optimal controls is challenging in particular for quantum systems that interact with their enviroment through noise and measurement (open quantum systems).Current quantum control methods are either open loop or make strong approximations of linear dynamics and Gaussian noise (LQG). Here we propose a new sets of stochastic optimal control methods to compute feedback controllers for open quantum systems, based on path integral methods.These methods can deal with system non-linearities and noise in a principled way.PI control has been applied with great success for classical systems such as robots but sofar not to quantum systems.While path integral methods for stochastic optimal control for classical problems have been well studied by a growing research community, their application to quantum problems is unexplored and provides a number of unique challenges. Design of feedback controllers requires measurement, which only provides partial information of the quantum state, making it a partially observed control problem. Control of this partial observable control problem is difficult in general, requiring an integrated treatment of the quantum state estimation and control computation. In this talk, I will outline the theory and give our first results on some challenging quantum state preparation problems.