Vehicle Behavior Recognition Method

Abstract

The invention relates to the field of vehicle intelligent perception, and discloses a vehicle behavior recognition method. The method includes: acquiring the current lateral distance of the preceding vehicle relative to the host vehicle and the first duration of the lateral distance change of the preceding vehicle; wherein, the preceding vehicle is a vehicle driving in front of and immediately adjacent to the host vehicle, and the The preceding vehicle and the host vehicle are driving in different lanes; based on the current lateral distance and the first duration, determine the first lane-changing two-degree-of-freedom chi-square value of the preceding vehicle, and based on the first The two-degree-of-freedom chi-square value of lane change determines the first lane-change probability of the preceding vehicle relative to the lane where the host vehicle is located; based on the first lane-changing probability and the first set lane-changing probability threshold, the The lane-changing behavior of the vehicle. The chi-square distribution of two degrees of freedom is used to identify the cut-in and cut-out behaviors of vehicles, and the efficiency of vehicle behavior recognition is improved on the basis of ensuring the accuracy of vehicle behavior recognition.

Description

technical field [0001] The invention relates to the field of vehicle intelligent perception, in particular to a vehicle behavior recognition method. Background technique [0002] For intelligent driving systems such as automatic driving and assisted driving, it mainly makes decisions based on the real-time status information of the surrounding objects provided by various sensors on the vehicle. The system cannot predict the upcoming behavior in advance, especially for the front vehicle. In extremely dangerous scenarios such as sudden cut-in and sudden deceleration at close range, the system will be greatly restricted. Based on this, some methods of automatic recognition of vehicle behavior have emerged in order to predict the upcoming behavior of the vehicle. [0003] However, the current vehicle behavior recognition methods are mainly implemented based on deep learning algorithms, which not only need to collect a large amount of data to train the model, but also require th...

AI Technical Summary

Problems solved by technology

[0003] However, the current vehicle behavior recognition methods are mainly implemented based on deep learning algorithms, which not only need to collect a large amount of data to train the model, but also require the system to have strong computing power, which will result in high system implementation costs and also due to The model is more complex and affects the prediction efficiency of vehicle behavior

Images