This paper proposes a robust optimization (RO) model for bidirectional dispatch coordination of large-scale plug-in electric vehicles (PEVs) in a power grid in which the PEVs are aggregated to manage. The PEV aggregators are considered as a type of dispatchable demand response and energy storage resource with stochastic behaviors, and can supply load or provide ancillary services such as regulation reserve to the grid. The proposed RO model is then reformulated as a mixed-integer quadratic programming model, which can be solved efficiently. Computer simulations are performed for a power grid with ten generators and three PEV aggregators to validate the economic benefit of the RO model for bidirectional dispatch coordination of the PEVs and the robustness of the RO model to the uncertainty of the PEVs’ stochastic mobility behaviors.