Vehicle driving control method, device and equipment, vehicle and medium

Abstract

The invention relates to a vehicle driving control method, device and equipment, a vehicle and a medium in the field of automatic driving. The method comprises the steps that vehicle driving data of the vehicle in the driving process is acquired; the vehicle driving data are input into a driving analysis model obtained through pre-training to obtain target driving information output by the driving analysis model, and the driving analysis model is obtained by screening historical driving sample data and training an initial analysis model through the target driving sample data obtained through screening; the target driving sample data is data with confidence lower than a confidence threshold in the historical driving sample data; and controlling vehicle driving according to the target driving information. By screening the target driving sample data with the confidence lower than the confidence threshold from the historical driving sample data to train the initial analysis model, the recall rate, accuracy and other model performances of the driving analysis model obtained by training can be improved, and the accuracy and safety of vehicle driving control are further improved.

Description

technical field [0001] The present disclosure relates to the technical field of vehicle engineering, and in particular, to a vehicle driving control method, device, device, vehicle and medium. Background technique [0002] During the driving process of the vehicle, it is usually necessary to process the collected driving data of the vehicle by means of the model, so as to determine the driving control strategy of the vehicle according to the output of the model. When training the model, it is usually necessary to label a large amount of historical road collection data, and input the marked historical road collection data into the neural network model to complete the training of the model. However, due to the repetition of the collected historical road collection data Scenes not only lead to a large workload of labeling, but also the accuracy of the model trained as training samples is low through the low-value road-collection data such as these repeated scenes after labeling...

AI Technical Summary

Problems solved by technology

When training the model, it is usually necessary to label a large amount of historical road mining data, and input the marked historical road mining data into the neural network model to complete the training of the model. Scenes, not only lead to a large workload of labeling, but also low-value road sampling data such as these repeated scenes after labeling, as training samples, the accuracy of the model obtained by training is low

Images