Method for pre-judging collision possibility of unmanned vehicle and surrounding vehicles

Abstract

The invention relates to the technical field of unmanned driving, and provides a method for pre-judging the collision possibility of an unmanned vehicle and surrounding vehicles. The method comprisesthe following steps of firstly, track data and environment data of a target vehicle are acquired; feature value addition and standardization processing are carried out on the trajectory data, lane changing and straight-going trajectory data are extracted and marked, and a kLSTM data set is constructed; environment data is cleaned, lane changing and straight-movingenvironment data is extracted andlabelled, and an Adaboost data set is created; a kAttention-LSTM deep neural network model and an Adaboost machine learning model are created and trained and behavior pre-judgment is made on the target vehicle in combination with the three models; after the lane changing behavior is predicted, collision possibility pre-judgment is conducted, and collision avoidance measures are taken when it is judged that the vehicle is in the dangerous driving state. According to the invention, the accuracy, real-time performance and efficiency of pre-judging the collision possibility of the unmanned vehicleand the surrounding vehicles can be improved.

Description

technical field [0001] The invention relates to the technical field of unmanned driving, in particular to a method for predicting the possibility of collision between an unmanned vehicle and surrounding vehicles. Background technique [0002] With the continuous increase in the incidence of global traffic accidents and the continuous increase of car ownership, traffic congestion and traffic safety problems are becoming more and more serious, and gradually arouse people's widespread attention. According to statistics, from 2005 to 2018, more than 1.25 million people died in road traffic accidents every year around the world, and China ranked first, the country with the highest number of road traffic accident deaths in the world. Moreover, with the steady development of China's economy and the continuous improvement of people's living standards, the number of cars in my country continues to rise. According to statistics released by the public security traffic control departme...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems of low accuracy, poor real-time performance, overly complicated methods, and low efficiency existing in the prior art, the present invention provides a method for predicting the possibility of collision between an unmanned vehicle and surrounding vehicles, which can improve the collision probability between an unmanned vehicle and surrounding vehicles. Accuracy, real-time and efficiency of vehicle collision possibility prediction

Images