Pedestrian intention prediction data closed loop method, apparatus and device, and storage medium

The invention provides a pedestrian intention prediction data closed-loop method. The method comprises the following steps: acquiring target scene data and pedestrian trajectory data; classifying and marking the pedestrian trajectory data to obtain a pedestrian intention sample data set; performing model training on the pedestrian intention sample data set to obtain a pedestrian intention prediction result; acquiring real-time scene data and real-time pedestrian trajectory data of the target vehicle, and classifying and marking the real-time pedestrian trajectory data to obtain a real vehicle data set; comparing and evaluating the pedestrian intention prediction result with the real vehicle data set to obtain a model evaluation result; and real-time pedestrian trajectory data is classified and labeled and then input into the model for supplementary training, so that a data closed loop is formed. The invention further provides a device, equipment and a storage medium. In the scheme of the invention, the real v