Ethical Decision-Oriented Research On Urban Traffic Scenes Based On Object Recognition Algorithm

Although machine ethics has been developed for decades,with the rise of artificial intelligence,artificial intelligence ethics has replaced machine ethics and become a new challenge in the development of information technology.Intelligent driving,as one of the applications of artificial intelligence closest to people’s lives,has attracted wide attention to the risks and challenges of ethical decision-making brought by its driving behavior.Academia,policy makers and automobile manufacturers have been trying to deal with the ethical problems brought by intelligent driving vehicles in the development process of intelligent driving.At present,most studies on the ethical issues of intelligent driving by experts and scholars are mainly conducted by online research.Although this method can collect more information related to people’s wishes,it cannot obtain enough road information to assist intelligent driving cars to make decisions.Therefore,this paper proposes a method to analyze urban road scenes by using object recognition technology,so as to provide technical support for solving ethical problems in intelligent driving.The main research contents of this paper are as follows:1.Urban road traffic elements Extraction for ethical decision-making of intelligent driving.This paper mainly studies the ethical issues of intelligent driving vehicles.For this purpose,it describes in detail the traffic scenes that may have ethical problems in urban roads,and extracts the main traffic elements related to the ethical issues of intelligent driving in the scene,including pedestrians,riders,vehicles and species.Furthermore,the ethical attributes related to these four categories of traffic elements are extracted,and 23 category labels related to traffic elements are redefined according to these ethical attributes.2.Construction of urban road scene data set for traffic element identification.In order to enable intelligent driving vehicles to identify traffic elements with ethical attributes during driving,this paper constructs a new image data set based on the redefined category labels with ethical attributes.A large number of experiments based on this image data set prove that intelligent driving vehicles can use computer vision technology to identify traffic elements with ethical attribute labels,which provides a technical basis for more direct research on ethical issues in intelligent driving.3.Research on ethical decision-making methods of intelligent driving in line with ethics.In order to train intelligent driving vehicles to be equipped with the ability to deal with ethical dilemmas contained in urban road scenes,this paper analyzes the ethical dilemmas of intelligent driving vehicles in urban road scenes.Combining the ethical attribute labels in the data set of this paper and a large number of survey data collected by previous generations,It provides an ethical decision-making method of driving strategy for intelligent driving vehicles in ethical dilemma.