Vehicle behavior data labeling method and device

Abstract

The embodiment of the invention discloses a vehicle behavior data labeling method and device. The method comprises the steps of obtaining to-be-labeled first vehicle behavior data including vehicle information and road information, calculating statistical characteristics of the vehicle information based on a pre-established target mathematical statistical model, and determining a target statistical rule satisfied by the statistical characteristics, obtaining behavior types corresponding to the first vehicle behavior data according to the target statistical rule and the road information, and when the accuracy of the behavior types corresponding to all the first vehicle behavior data in the obtained target data set is greater than a preset threshold value, determining that the first vehiclebehavior data is abnormal; marking the first vehicle behavior data based on the obtained behavior type to obtain corresponding marking information, and when the obtained accuracy is not greater than apreset threshold, marking second vehicle behavior data in a new target data set based on a pre-established target network model to obtain corresponding marking information. By applying the scheme provided by the embodiment of the invention, labeling can be automatically carried out, and the labeling efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of intelligent driving, in particular to a method and device for labeling vehicle behavior data. Background technique [0002] In the autonomous driving scenario, predicting the behavior of other cars in advance and issuing warnings can avoid traffic accidents. Therefore, the future behavior prediction of other cars is very necessary. [0003] At present, the behavior of other cars is mainly predicted through machine learning. In order to establish a behavior prediction machine learning model, a large amount of vehicle behavior data and labeling information after labeling the vehicle behavior data are required. For example, when the vehicle information in the vehicle behavior data is the left turn signal flashing, and the road information is the left turn lane , the label information can be that the vehicle turns left. [0004] At present, domestic and foreign data labeling work mostly adopts manua...

AI Technical Summary

Problems solved by technology

The above method needs to employ a large number of human resources to meet the labeling requirements in the continuous iteration process of machine learning, which is inefficien

Images