In this talk, we consider high-dimensional traffic signal control problems that arise in congested metropolitan areas. We focus on the use of high-resolution urban mobility stochastic simulators and formulate the control problems as high-dimensional continuous simulation-based optimization (SO) problems. We discuss the opportunities and challenges of designing SO algorithms for these problems. An important component in high-dimensional problems is the exploration-exploitation tradeoff. We discuss work that has focused on improving the exploitation capabilities of SO algorithms. We then present novel exploration techniques suitable for high-dimensional spaces. We consider a Bayesian optimization setting, and propose the use of a simple analytical traffic model to specify the covariance function of a Gaussian process. We show how this enables the Bayesian optimization method to more efficiently sample in high-dimensional spaces. We present validation experiments on synthetic low-dimensional problems. We then apply the method to a high-dimensional traffic control problem for Midtown Manhattan, in New York City.