Simulation Of Virtual-real Hybrid Traffic Scene Based On Digital Twin

With the acceleration of my country’s urbanization process and the rapid growth of car ownership,traffic problems in modern cities have become more and more serious.Traffic safety and efficiency issues have become the main challenges faced by urban traffic.Real traffic scenes have great difficulties in providing repeatable and controllable traffic scenes for related research,and it is impossible to fully simulate traffic conditions in different traffic scenes.At the same time,traffic scene simulation experiments on open roads will face a series of problems of safety assurance and high cost.Therefore,closed field tests have become a common means of traffic scene simulation experiments.However,closed field tests also have traffic scenes,traffic events,limit The problem of high repetitive costs in traffic scenes.In order to partially solve the above problems,based on the sand table of the multi-mode intelligent networked traffic miniature simulation platform of Chang’an University,combined with the digital twin technology,this paper proposes a simulation system of virtual-real hybrid traffic scene based on digital twin,aiming to solve the problems of traditional traffic scene simulation methods.issues of functionality,safety,and cost.Although there is a gap with real traffic scenes,the construction of traffic scenes based on digital twins provides a safe,controllable,repeatable,and low-cost traffic scenario simulation method,especially for extreme traffic scenes.This paper has done the following two aspects of work.(1)Based on the school’s multi-mode intelligent network traffic miniature simulation platform sand table,firstly a method based on Unity3 D simulation of virtual traffic scenes is proposed,the simulation platform is modeled,and vehicle driving environments in different climates and locations are simulated.Secondly,through the virtual-real mapping and collision detection in the digital twin technology,the data connection and state interaction between the physical vehicle and the virtual vehicle are realized.Finally,the interaction design of the virtual driving system is completed for the visualization and manipulation of the vehicle.The driving simulation allows the driver to control the vehicle from a first-person or third-person perspective,while also being assisted by artificial intelligence.Experiments were carried out in different virtual traffic scenes,which solved the problem that the extreme traffic scenes in the simulation experiments of real traffic scenes were difficult to complete and the cost was high.(2)The simulation system of virtual-real hybrid traffic scene completes the simulation of car-following scenes,vehicle lane-changing scenes,and vehicle collision scenes by combining the "master vehicle" driven by humans and the "autonomous vehicle" with automatic driving functions.The master vehicle can controllably repeat three types of traffic scenes according to the test needs.The system can be used to test autonomous driving algorithms and is suitable for scenes where physical vehicles are limited.Experimental results prove that the simulation system of virtual-real hybrid traffic scene based on digital twin proposed in this paper can simulate the interaction characteristics between real vehicles and virtual autonomous vehicles.This system can simulate a variety of traffic scenes,save data collection costs,and is expected to be used as a reference in future research on automatic driving systems.