The invention provides a data-driven path planning method for an unmanned underwater vehicle, and is mainly applied to path planning of an underwater region traversal task. The method comprises the following steps: firstly, for a task area, carrying out environment modeling by utilizing a grid method, reading priori knowledge to update a modeling map, and generating an initial global path by utilizing initial map information; during execution of the task, detecting and receiving observation data in real time, carrying out online path planning analysis, deciding a global planning mode and a local planning mode based on a planning mode decision evaluation function F, if the global planning mode and the local planning mode are switched to the local planning mode, carrying out local planning according to a heuristic rule set, and returning to an initial global planning path after traversal of a local feature dense area is completed; according to the scheme, online guidance decision makingis carried out according to underwater real-time high-dimensional space observation data, global and local planning decision making is carried out on the basis of real-time detection conditions, an adaptive planning strategy is selected through planning mode switching, the method can be suitable for more complex environments, and the time efficiency is remarkably improved.