Human attention experience learning method and device for automatic driving of vehicle

The invention discloses a human attention experience learning method and device for vehicle automatic driving, and the method comprises the steps: obtaining current driving scene data which comprises at least one piece of regional data; processing the current driving scene data through an attention distribution model to obtain weight values of different regional data; and processing the regional data in sequence from the high weight value to the low weight value, and generating a driving decision. After the current driving scene data of the vehicle is obtained, weight values of different area data in the current driving scene data are obtained through an attention distribution model, namely, the importance degree of each area data in the current driving scene data in different driving environments can be distinguished by the automatic driving vehicle; and area data with high weight values are processed preferentially, that is, the automatic driving system focuses on processing of key information with higher w