To be properly controlled, dynamic systems need plans or policies. Plans are sequences of actions to be performed, whereas policies associate an action to be performed with each possible system state. The model-based synthesis of plans or policies consists in producing them automatically starting from a model of the physical system to be controlled and from user requirements on the controlled system. This article is a survey of what exists and what has been done at Onera for the automatic synthesis of plans or policies for the high-level control of dynamic systems.