Research On The Acceptance Of Independent And Cooperative Vehicle-highway Autonomous Driving Technology

Autonomous driving is an important part of intelligent transportation construction.The increase in the proportion of autonomous vehicles in cities is expected to alleviate traffic congestion,improve land use structure,and increase travel efficiency.Various impacts and changes caused by autonomous driving technology are closely related to public acceptance.Independent autonomous driving and cooperative vehicle-highway autonomous driving may be the two development directions of autonomous vehicles in the future,but few studies compare the public acceptance of the two.This paper proposes a combined structural equation model and artificial neural network method to compare and analyze the public acceptance of independent autonomous driving and cooperative vehicle-highway autonomous driving.The structural equation model(SEM)is frequently used to study multiple predictors of autonomous vehicles acceptance and analyze the linear relationship between factors.The artificial neural network(ANN)can explain nonlinear correlation and express the weight of influence factors.The combination of them not only makes up for the lack of linear analysis of structural equation model,but also ensures the input accuracy of artificial neural network.After comparing the influencing factors of acceptance and use of autonomous driving in the existing research,the extended Unified Theory of Acceptance and Use of Technology(UTAUT)is used to guide the theoretical framework of the whole study.The motivational factors,utilization of resources and external norms that influence people's decision-making are covered by four original variables in the UTAUT model,namely performance expectation,effort expectation,social influence and facilitating conditions.The two variables of expanded risk expectation and consumer innovation enrich the theoretical framework from the external conditions of technology use and people's psychology,which comprehensively explains the acceptance of autonomous driving.Use the structural equation model to analyze the causal relationship between the dependent variables and independent variables and the linear correlation between latent variables.We systematically study the influencing factors of the acceptance of autonomous driving through regression analysis,and the significance of each variable in the prediction of acceptance intention is verified.The six factors with significant predictive power verified by the structural equation model are used as the input layer of the artificial neural network,and the output layer is the accept intention.In order to quantify the predictive relationship between input variables and output variable,an artificial neural network model is structed in the feedforward back propagation multilayer perceptron.We find that the six variables of the theoretical framework have significant effects on the acceptance of autonomous driving,social influence is the strongest predictor of the acceptance of independent autonomous driving,and the most significant factor of the acceptance of cooperative vehicle-highway autonomous driving is effort expectation.Additionally,the neural network analysis of the prediction factors of risk expectation and performance expectation is carried out,and different weights of input neurons are found.This study tries to adopt a systematic theoretical framework to cover multiple factors affecting the acceptance of autonomous driving of heterogeneous population.This paper explores the relationship between the significance and relative importance of complex variables by the analysis method of structural equation model and artificial neural network.We aim to provide technology developers and industry managers with relevant theoretical foundations for autonomous driving technology development and policy formulation.